52
June 23, 2015 Semantic Web Meetup, SEO San Diego Meetup, SEM San Diego Meetup Courtyard San Diego Old Town #SEOSDM

Ranking in Google Since The Advent of The Knowledge Graph

Embed Size (px)

Citation preview

June 23 2015 Semantic Web Meetup SEO San Diego Meetup SEM San Diego Meetup Courtyard San Diego Old Town

SEOSDM

A Two Person Panel DiscussionPresentation by Bill Slawski and Barbara Starr

User experience drives search engines and hence their results Search Engine Result PresentationPlacements (SERPs) naturally follow that route This means that search results are no longer exclusively based on just ranking criteria Amongst other critical factors is understanding the notion of ordering vs ranking the impact of context and many others

Search Engine Results Page

Search Engine Results Placement

bill_slawski amp BarbaraStarr

Ranking search results based on entity metrics

Providing Knowledge Panels With Search Results

Maintaining search context

Near-duplicate filtering in search engine result page of an online shopping system

Clustered search results

ldquoReturning by one or more computing devices an ordered list of results responsive to the query from the data store of an online shopping system filtered as a function of at least one of the distance and the cluster identifierrdquo

Near-duplicate filtering in search engine result page of an online shopping system

ldquoA plurality of metrics is determined associated with a search result obtained from a knowledge graph wherein the metrics are indicative of the relevance of the search result and the metrics are determined at least in part from the knowledge graph ldquo ldquoRelates to ranking search results Conventional techniques for ranking search results include alphabetical ordering and keyword matching ldquo ldquoResults may be image thumbnail links ordered horizontally based on scorerdquo Example metrics may be Notable Entity Type Metric Contribution Metric (and Fame Metric) Relatedness Metric Prize Metric

Ranking search results based on entity metrics

Notability Notable type Notable Type Metrics and more

US20130110825

Expertise in Entities

Meta-Web patent mentions a ldquobuy buttonrdquo 10 times ndash it was a commercial startup

bill_slawski amp BarbaraStarr

Automated online purchasing system

Meta-Web

Delegated authority evaluation system

User Contributed Knowledge Database

Graph Store

Knowledge web

Meta-Web Search Results

US20040210602A1

ldquoGenerating a buy button with which the user can enter into a personalized purchase transaction to bring the user to a preferred vendor or list of vendorsrdquo

rdquoThe search results page further includes one or more items that when selected by the user lead to a product node for a particular product ldquo

Meta-Web

ldquothe registry establishes connections between objects stored in the registry the connections comprising typed lines between the registry objects

US201450100569A1

US20150100569A1

In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools The knowledge may be in the form of documents or other media or it may be a descriptor of a book or other physical source

Knowledge web

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

A Two Person Panel DiscussionPresentation by Bill Slawski and Barbara Starr

User experience drives search engines and hence their results Search Engine Result PresentationPlacements (SERPs) naturally follow that route This means that search results are no longer exclusively based on just ranking criteria Amongst other critical factors is understanding the notion of ordering vs ranking the impact of context and many others

Search Engine Results Page

Search Engine Results Placement

bill_slawski amp BarbaraStarr

Ranking search results based on entity metrics

Providing Knowledge Panels With Search Results

Maintaining search context

Near-duplicate filtering in search engine result page of an online shopping system

Clustered search results

ldquoReturning by one or more computing devices an ordered list of results responsive to the query from the data store of an online shopping system filtered as a function of at least one of the distance and the cluster identifierrdquo

Near-duplicate filtering in search engine result page of an online shopping system

ldquoA plurality of metrics is determined associated with a search result obtained from a knowledge graph wherein the metrics are indicative of the relevance of the search result and the metrics are determined at least in part from the knowledge graph ldquo ldquoRelates to ranking search results Conventional techniques for ranking search results include alphabetical ordering and keyword matching ldquo ldquoResults may be image thumbnail links ordered horizontally based on scorerdquo Example metrics may be Notable Entity Type Metric Contribution Metric (and Fame Metric) Relatedness Metric Prize Metric

Ranking search results based on entity metrics

Notability Notable type Notable Type Metrics and more

US20130110825

Expertise in Entities

Meta-Web patent mentions a ldquobuy buttonrdquo 10 times ndash it was a commercial startup

bill_slawski amp BarbaraStarr

Automated online purchasing system

Meta-Web

Delegated authority evaluation system

User Contributed Knowledge Database

Graph Store

Knowledge web

Meta-Web Search Results

US20040210602A1

ldquoGenerating a buy button with which the user can enter into a personalized purchase transaction to bring the user to a preferred vendor or list of vendorsrdquo

rdquoThe search results page further includes one or more items that when selected by the user lead to a product node for a particular product ldquo

Meta-Web

ldquothe registry establishes connections between objects stored in the registry the connections comprising typed lines between the registry objects

US201450100569A1

US20150100569A1

In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools The knowledge may be in the form of documents or other media or it may be a descriptor of a book or other physical source

Knowledge web

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Search Engine Results Page

Search Engine Results Placement

bill_slawski amp BarbaraStarr

Ranking search results based on entity metrics

Providing Knowledge Panels With Search Results

Maintaining search context

Near-duplicate filtering in search engine result page of an online shopping system

Clustered search results

ldquoReturning by one or more computing devices an ordered list of results responsive to the query from the data store of an online shopping system filtered as a function of at least one of the distance and the cluster identifierrdquo

Near-duplicate filtering in search engine result page of an online shopping system

ldquoA plurality of metrics is determined associated with a search result obtained from a knowledge graph wherein the metrics are indicative of the relevance of the search result and the metrics are determined at least in part from the knowledge graph ldquo ldquoRelates to ranking search results Conventional techniques for ranking search results include alphabetical ordering and keyword matching ldquo ldquoResults may be image thumbnail links ordered horizontally based on scorerdquo Example metrics may be Notable Entity Type Metric Contribution Metric (and Fame Metric) Relatedness Metric Prize Metric

Ranking search results based on entity metrics

Notability Notable type Notable Type Metrics and more

US20130110825

Expertise in Entities

Meta-Web patent mentions a ldquobuy buttonrdquo 10 times ndash it was a commercial startup

bill_slawski amp BarbaraStarr

Automated online purchasing system

Meta-Web

Delegated authority evaluation system

User Contributed Knowledge Database

Graph Store

Knowledge web

Meta-Web Search Results

US20040210602A1

ldquoGenerating a buy button with which the user can enter into a personalized purchase transaction to bring the user to a preferred vendor or list of vendorsrdquo

rdquoThe search results page further includes one or more items that when selected by the user lead to a product node for a particular product ldquo

Meta-Web

ldquothe registry establishes connections between objects stored in the registry the connections comprising typed lines between the registry objects

US201450100569A1

US20150100569A1

In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools The knowledge may be in the form of documents or other media or it may be a descriptor of a book or other physical source

Knowledge web

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Ranking search results based on entity metrics

Providing Knowledge Panels With Search Results

Maintaining search context

Near-duplicate filtering in search engine result page of an online shopping system

Clustered search results

ldquoReturning by one or more computing devices an ordered list of results responsive to the query from the data store of an online shopping system filtered as a function of at least one of the distance and the cluster identifierrdquo

Near-duplicate filtering in search engine result page of an online shopping system

ldquoA plurality of metrics is determined associated with a search result obtained from a knowledge graph wherein the metrics are indicative of the relevance of the search result and the metrics are determined at least in part from the knowledge graph ldquo ldquoRelates to ranking search results Conventional techniques for ranking search results include alphabetical ordering and keyword matching ldquo ldquoResults may be image thumbnail links ordered horizontally based on scorerdquo Example metrics may be Notable Entity Type Metric Contribution Metric (and Fame Metric) Relatedness Metric Prize Metric

Ranking search results based on entity metrics

Notability Notable type Notable Type Metrics and more

US20130110825

Expertise in Entities

Meta-Web patent mentions a ldquobuy buttonrdquo 10 times ndash it was a commercial startup

bill_slawski amp BarbaraStarr

Automated online purchasing system

Meta-Web

Delegated authority evaluation system

User Contributed Knowledge Database

Graph Store

Knowledge web

Meta-Web Search Results

US20040210602A1

ldquoGenerating a buy button with which the user can enter into a personalized purchase transaction to bring the user to a preferred vendor or list of vendorsrdquo

rdquoThe search results page further includes one or more items that when selected by the user lead to a product node for a particular product ldquo

Meta-Web

ldquothe registry establishes connections between objects stored in the registry the connections comprising typed lines between the registry objects

US201450100569A1

US20150100569A1

In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools The knowledge may be in the form of documents or other media or it may be a descriptor of a book or other physical source

Knowledge web

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

ldquoReturning by one or more computing devices an ordered list of results responsive to the query from the data store of an online shopping system filtered as a function of at least one of the distance and the cluster identifierrdquo

Near-duplicate filtering in search engine result page of an online shopping system

ldquoA plurality of metrics is determined associated with a search result obtained from a knowledge graph wherein the metrics are indicative of the relevance of the search result and the metrics are determined at least in part from the knowledge graph ldquo ldquoRelates to ranking search results Conventional techniques for ranking search results include alphabetical ordering and keyword matching ldquo ldquoResults may be image thumbnail links ordered horizontally based on scorerdquo Example metrics may be Notable Entity Type Metric Contribution Metric (and Fame Metric) Relatedness Metric Prize Metric

Ranking search results based on entity metrics

Notability Notable type Notable Type Metrics and more

US20130110825

Expertise in Entities

Meta-Web patent mentions a ldquobuy buttonrdquo 10 times ndash it was a commercial startup

bill_slawski amp BarbaraStarr

Automated online purchasing system

Meta-Web

Delegated authority evaluation system

User Contributed Knowledge Database

Graph Store

Knowledge web

Meta-Web Search Results

US20040210602A1

ldquoGenerating a buy button with which the user can enter into a personalized purchase transaction to bring the user to a preferred vendor or list of vendorsrdquo

rdquoThe search results page further includes one or more items that when selected by the user lead to a product node for a particular product ldquo

Meta-Web

ldquothe registry establishes connections between objects stored in the registry the connections comprising typed lines between the registry objects

US201450100569A1

US20150100569A1

In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools The knowledge may be in the form of documents or other media or it may be a descriptor of a book or other physical source

Knowledge web

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

ldquoA plurality of metrics is determined associated with a search result obtained from a knowledge graph wherein the metrics are indicative of the relevance of the search result and the metrics are determined at least in part from the knowledge graph ldquo ldquoRelates to ranking search results Conventional techniques for ranking search results include alphabetical ordering and keyword matching ldquo ldquoResults may be image thumbnail links ordered horizontally based on scorerdquo Example metrics may be Notable Entity Type Metric Contribution Metric (and Fame Metric) Relatedness Metric Prize Metric

Ranking search results based on entity metrics

Notability Notable type Notable Type Metrics and more

US20130110825

Expertise in Entities

Meta-Web patent mentions a ldquobuy buttonrdquo 10 times ndash it was a commercial startup

bill_slawski amp BarbaraStarr

Automated online purchasing system

Meta-Web

Delegated authority evaluation system

User Contributed Knowledge Database

Graph Store

Knowledge web

Meta-Web Search Results

US20040210602A1

ldquoGenerating a buy button with which the user can enter into a personalized purchase transaction to bring the user to a preferred vendor or list of vendorsrdquo

rdquoThe search results page further includes one or more items that when selected by the user lead to a product node for a particular product ldquo

Meta-Web

ldquothe registry establishes connections between objects stored in the registry the connections comprising typed lines between the registry objects

US201450100569A1

US20150100569A1

In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools The knowledge may be in the form of documents or other media or it may be a descriptor of a book or other physical source

Knowledge web

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Notability Notable type Notable Type Metrics and more

US20130110825

Expertise in Entities

Meta-Web patent mentions a ldquobuy buttonrdquo 10 times ndash it was a commercial startup

bill_slawski amp BarbaraStarr

Automated online purchasing system

Meta-Web

Delegated authority evaluation system

User Contributed Knowledge Database

Graph Store

Knowledge web

Meta-Web Search Results

US20040210602A1

ldquoGenerating a buy button with which the user can enter into a personalized purchase transaction to bring the user to a preferred vendor or list of vendorsrdquo

rdquoThe search results page further includes one or more items that when selected by the user lead to a product node for a particular product ldquo

Meta-Web

ldquothe registry establishes connections between objects stored in the registry the connections comprising typed lines between the registry objects

US201450100569A1

US20150100569A1

In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools The knowledge may be in the form of documents or other media or it may be a descriptor of a book or other physical source

Knowledge web

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

US20130110825

Expertise in Entities

Meta-Web patent mentions a ldquobuy buttonrdquo 10 times ndash it was a commercial startup

bill_slawski amp BarbaraStarr

Automated online purchasing system

Meta-Web

Delegated authority evaluation system

User Contributed Knowledge Database

Graph Store

Knowledge web

Meta-Web Search Results

US20040210602A1

ldquoGenerating a buy button with which the user can enter into a personalized purchase transaction to bring the user to a preferred vendor or list of vendorsrdquo

rdquoThe search results page further includes one or more items that when selected by the user lead to a product node for a particular product ldquo

Meta-Web

ldquothe registry establishes connections between objects stored in the registry the connections comprising typed lines between the registry objects

US201450100569A1

US20150100569A1

In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools The knowledge may be in the form of documents or other media or it may be a descriptor of a book or other physical source

Knowledge web

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Expertise in Entities

Meta-Web patent mentions a ldquobuy buttonrdquo 10 times ndash it was a commercial startup

bill_slawski amp BarbaraStarr

Automated online purchasing system

Meta-Web

Delegated authority evaluation system

User Contributed Knowledge Database

Graph Store

Knowledge web

Meta-Web Search Results

US20040210602A1

ldquoGenerating a buy button with which the user can enter into a personalized purchase transaction to bring the user to a preferred vendor or list of vendorsrdquo

rdquoThe search results page further includes one or more items that when selected by the user lead to a product node for a particular product ldquo

Meta-Web

ldquothe registry establishes connections between objects stored in the registry the connections comprising typed lines between the registry objects

US201450100569A1

US20150100569A1

In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools The knowledge may be in the form of documents or other media or it may be a descriptor of a book or other physical source

Knowledge web

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Automated online purchasing system

Meta-Web

Delegated authority evaluation system

User Contributed Knowledge Database

Graph Store

Knowledge web

Meta-Web Search Results

US20040210602A1

ldquoGenerating a buy button with which the user can enter into a personalized purchase transaction to bring the user to a preferred vendor or list of vendorsrdquo

rdquoThe search results page further includes one or more items that when selected by the user lead to a product node for a particular product ldquo

Meta-Web

ldquothe registry establishes connections between objects stored in the registry the connections comprising typed lines between the registry objects

US201450100569A1

US20150100569A1

In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools The knowledge may be in the form of documents or other media or it may be a descriptor of a book or other physical source

Knowledge web

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Meta-Web Search Results

US20040210602A1

ldquoGenerating a buy button with which the user can enter into a personalized purchase transaction to bring the user to a preferred vendor or list of vendorsrdquo

rdquoThe search results page further includes one or more items that when selected by the user lead to a product node for a particular product ldquo

Meta-Web

ldquothe registry establishes connections between objects stored in the registry the connections comprising typed lines between the registry objects

US201450100569A1

US20150100569A1

In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools The knowledge may be in the form of documents or other media or it may be a descriptor of a book or other physical source

Knowledge web

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

ldquoGenerating a buy button with which the user can enter into a personalized purchase transaction to bring the user to a preferred vendor or list of vendorsrdquo

rdquoThe search results page further includes one or more items that when selected by the user lead to a product node for a particular product ldquo

Meta-Web

ldquothe registry establishes connections between objects stored in the registry the connections comprising typed lines between the registry objects

US201450100569A1

US20150100569A1

In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools The knowledge may be in the form of documents or other media or it may be a descriptor of a book or other physical source

Knowledge web

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

US201450100569A1

US20150100569A1

In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools The knowledge may be in the form of documents or other media or it may be a descriptor of a book or other physical source

Knowledge web

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools The knowledge may be in the form of documents or other media or it may be a descriptor of a book or other physical source

Knowledge web

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Brand Identifiers

Entity Identifiers

bill_slawski amp BarbaraStarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Providing search results based on a compositional query

Crowdsourcing user-provided identifiers and associating them with brand identities

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Entity Identifier

WO2014089769A1

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Brand Identifier

US20140250192A1

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

ldquo In some implementations search results include results identifying entity references As used herein an entity reference is an identifier eg text or other information that refers to an entity For example an entity may be the physical embodiment of George Washington while an entity reference is an abstract concept that refers to George Washington Where appropriate based on context it will be understood that the term entity as used herein may correspond to an entity reference and the term entity reference as used herein may correspond to an entity In some implementations the search system may identify an entity type associated with an entity reference The entity type may be a categorization or classification used to identify entity references in the data structure For example the entity reference George Washington may be associated with the entity types U S President ldquo Person and Military Officerrdquordquo

Providing search results based on a compositional query

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Different user-provided brand identifiers are extracted from messages provided by users of a social network The identifiers are aggregated into two or more aggregate identity groups When a brand identifier associated with a user request for content is determined to be in at least one of the aggregate identity groups content items comprising one or more other brand identifiers of the at least one aggregate identity group are provided to the user

Crowdsourcing user-provided identifiers And associating them with brand identities

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

In Freebase

Internally in the upcoming API bill_slawski amp BarbaraStarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Meta-Web

Query Optimization

Providing Search Results based on a Compositional Query

Question answering using entity references in unstructured data

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Query Optimization

US20100121839A1

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

ldquoentity references comprises ranking based on at least one ranking signal ldquo ldquoan entity result is selected from the one or more entity references based at least in part on the ranking An answer to the query is provided based at least in part on the entity resultrdquo

Question answering using entity references in unstructured data

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

ldquoWe describe the query optimization Techniques used by graphd a schema-last automatically indexed tuple-store which Supports freebasecom a world-writable database We demonstrate that the techniques described deliver performance that is generally comparable with traditional cost-based optimization techniques applied to the relational modelrdquo

Query Optimization

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

More Revenue (ads)

Action - Entity Pairs

bill_slawski amp BarbaraStarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Entity-based searching with content selection

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

System and method for providing contextual actions on a search results page

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

US20140258014

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

ldquoAnnotation describing a user interface that is to be visually displayed in connection with information identifying the document when the information identifying the document is included in a search results document the user interface including a user interface element that when selected causes an action to be performed in connection with the documentrdquo

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

ldquoRetrieving search results based in part on the search query identifying an entity-action pair comprising the named entity and an online action associated with the entity conducting a content auction for the entity-action pair based in part on auction bids received for the entity-action pairs selecting third-party content based on a result of the content auctionrdquo

Entity-based searching with content selection

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Standardize Entities

Contexts Structure (IOT)

bill_slawski amp BarbaraStarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

Providing entity-specific content in response to a search query (Microsoft)

Entity detection and extraction for entity cards (Microsoft)

Providing entity-specific content in response to a search query (Microsoft)

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

US20120059838A1

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

httpswwwseroundtablecomgoogle-mobile-color-lines-19898html

Mobile Card Interface

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

ldquoSeparate Templates may be used for separate factual entities In the case of a person the template may specify a description of the person and facts about the Person such as birthdate birth location career definitions and the likerdquo

Apparatus and Method for Supplying Search Results with a Knowledge Card (Unpublished Google Provisional Patent)

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

TemplatesCards are objects that know how to display (place) themselves Based on the device type

SERPS templates in this case are akin to ldquoresponsive designrdquo

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

As displayed in Blended Search Engine Results pages (SERPS)

Disparate data sourcessets mapped to distinct Search Engine Results Placements (SERPS)

bill_slawski amp BarbaraStarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Determination of a desired repository

Providing entity-specific content in response to a search query (Microsoft)

Interleaving search results

Browseable fact repository

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Browseable Fact Repository

US7774328B2

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Universal Search Repositories

US8266133B2

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

ldquoA system receives a search query from a user and searches a group of repositories based on the search query to identify for each of the repositories a set of search results The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identified repositoryrdquo

Determination of a desired repository

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

For Adwords Placement Panda traffic

For Data Quality

bill_slawski amp BarbaraStarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Classifying sites as low quality sites

Site quality score

Ranking search results

Predicting Site Quality

Providing a search results document that includes a user interface for performing an action in connection with a web page identified in the search results document

Focused Crawling for Structured Data (Paper)

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

ldquoA link quality score is determined for the site using the number of resources in each resource quality group If the link quality score is below a threshold link quality score the site is classified as a low quality siterdquo

Classifying sites as low quality sites

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

ldquoIn some implementations the search system identifies data in a data structure that includes quality scores Quality scores may be determined by global search history extracting scores from external websites search system developer input user preferences system settings predetermined parameters any other suitable technique or any combination thereof In an example the search system retrieves movie review scores from a website such as IMDB In another example the search system may retrieve restaurant reviews from YELP and a newspaper In some implementations multiple quality scores associated with an entity are combined in a weighted or unweighted techniquerdquo

Ranking search results based on entity metrics

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

ldquoWe propose new methods of focused crawling specifically designed for collecting data-rich pages with greater efficiency In particular we propose a novel combination of online learning and bandit-based exploreexploit approaches to predict data-rich web pages based on the context of the page as well as using feedback from the extraction of metadata from previously seen pages We show that these techniques significantly outperform state-of-the-art approaches for focused crawling measured as the ratio of relevant pages and non-relevant pages collected within a given budget rdquo

Focused Crawling For Structured Data

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr

Bill Slawski GoFishDigital Director of Search Marketing Editor SEO by the Sea httpsplusgooglecomu0+BillSlawski httpstwittercombill_slawski httpswwwlinkedincominslawski

Barbara Starr Semantic Fuse

Managing Partner and Founder httpsplusgooglecomu0+BarbaraStarr httpstwittercomBarbaraStarr httpswwwlinkedincominbarbarastarr