View
230
Download
0
Tags:
Embed Size (px)
Citation preview
INTERNATIONAL INSTITUTE FOR GEO-INFORMATION SCIENCE AND
EARTH OBSERVATION
Conceptualization of Place via Spatial
Clustering and Co-occurrence Analysis
2009 International Workshop on Location Based Social Networks (LBSN’09)
Dong–Po Deng; Tyng–Ruey Chuang; Rob Lemmens
Nov. 3, 2009, Seattle, WA, USA
GeoInformation is increasing on the Web
It’s a common activity for people to search and share geo-referenced information and resource on the Web
211/03/2009
From http://www.datenform.de/mapeng.html
Folksonomy
A tagging system allows users to classify objects of interests by keywords or terms
Folksonomy = practice of personal tagging of information and objects in social environment while people consume the information and use the objects
311/03/2009
Social tools
411/03/2009
Tags and Geo-tags
Tagging is a process that is established by keywords (k), users (u), and objects (o)
Geotag geo:lat=latitude e.g. geo:lat = 51.758 geo:lon=longitude e.g. geolong= 4.269€
Tag(k) ≡ ∀Keywords(k)
∧∃o(Object(o)∧associatedTo(k,o))
∧∃u(User(u)∧createdBy(k,u))
511/03/2009
Questions are …
Is geospatial data created in a social network a valuable production for a geospatial society in general?
How to extract the geospatial information from user-generated contents in a social network?
611/03/2009
Places as artifacts
Place is a center of meaning constructed by experiences
Place may be significant to any individual or group, and may exist at any scale
Locations become places only when activities occur that cause them to become imbued with meaning
Place provides the conditions of possibility for creative social practice
711/03/2009
Photos with tags = locations with tags
Tags
Tags
Tags
Tags
Collective intelligence
Tags should give rise to emergent semantics and shared conceptualization
Accumulation of tags on shared objects often express common consensus
Patterns and trends emerge from the collaboration and competition of many individuals are able to turn out structured information from tag-based system despite the lack of ontology and priori defined semantics
811/03/2009
911/03/2009
Photos and Tags in Flickr
Tags
Geo-Tag
Time-Tag
1011/03/2009
Selected photos from Flickr
1111/03/2009
Where is the beef?
2008 amsterdam canal europe holland netherlands noordholland north travel
The most frequently occurring 20%
1211/03/2009
Steps for extracting conceptualization of place
Tags
Tags
Tags
Tags
crawling
geotagged & tagged photos
database
Spatial clustering Co-occurrence analysis Place concepts
DBSCAN is a density-based algorithm
Two global parameters: Eps: Maximum radius of the neighbourhood MinPts: Minimum number of points in an Eps-
neighbourhood of that point Core Object: object with at least MinPts objects
within a radius ‘Eps-neighborhood’ Border Object: object that on the border of a
cluster
1311/03/2009
pq
MinPts = 5
Eps = 1 cm
Density-Based Clustering: Background
Density-reachable A point p is density-reachable
from a point q wrt Eps, MinPts if there is a chain of points p1, …, pn, p1 = q, pn = p such that pi+1 is directly density-reachable from pi
Density-connected A point p is density-connected to
a point q wrt. Eps, MinPts if there is a point o such that both, p and q are density-reachable from o wrt. Eps and MinPts.
1411/03/2009
p
qp1
p q
o
DBSCAN: The Algorithm
Arbitrary select a point p Retrieve all points density-reachable from p
wrt Eps and MinPts. If p is a core point, a cluster is formed. If p is a border point, no points are density-
reachable from p and DBSCAN visits the next point of the database.
Continue the process until all of the points have been processed.
1511/03/2009
1611/03/2009
Density-Based Clustering: Results
Co-occurrence analysis
Co-occurrence can be interpreted as an indicator of semantic similarity or an idiomatic expression.
Co-occurrence assumes interdependency of the two terms
Semantic similarity is a concept whereby a set of documents or terms within term lists are assigned a metric based on the likeness of their meaning / semantic content.
1711/03/2009
1811/03/2009
Co-occurrence matrix
The element at (i,j) is the tag count or frequency of the i’th tag in the j’th photos
jd
mnm
nTi
xx
xx
t
,1,
,11,1
1911/03/2009
Co-occurrence matrix
A row in the matrix is a vector of the tag’s occurrence in all photos:
While a column is a vector of the occurrence of all tags in a photo
][ ,1, niiTi xxt
jm
j
j
x
x
d
,
,1
2011/03/2009
Co-occurrence correlations
Photo-tag matrix
tag-tag correlation matrix
2111/03/2009
The correlation between the tag “amsterdam" and the tags of several landmarks associated to Amsterdam
Distance
Correlation coefficient
2211/03/2009
Conceptualizing places in 2500 meters
2311/03/2009
Conceptualizing places 150 meters
2411/03/2009
Conceptualizing places in 75 meters
Schiphol airport
2511/03/2009
Anne Frank House
2611/03/2009
Rijksmuseum
2711/03/2009
2811/03/2009
Conclusions and future works
Without the use of suitable spatial clustering, detailed information about a place is veiled by high frequency tags
A conceptualization of place is unveiled by tag co-occurrences at a suitable spatial scale
Location-based applications can be developed to suggest tags to users as they take photos
In the future we will ground the semantics between pairs of tags via the use of gazetteers or dictionaries
INTERNATIONAL INSTITUTE FOR GEO-INFORMATION SCIENCE AND
EARTH OBSERVATION
Thank you for your attention!
Dongpo [email protected]