Use cases for Graph Visualization
Corey Lanum & Sam Duby
The session will start at the top of the hour
16th August 2016
● Introductions
● Overview of use cases
● How to visualize graphs
○ Understanding terror networks
○ Tips for visualizing social networks
○ Detecting fraud with graphs
● Q&A - submit via the Citrix panel
Agenda
Reminder: This session is being recorded. Feel free to share!
Introductions
Sam Duby - Technical Sales EMEA
New to Cambridge Intelligence, looking after new &
existing customers in the EMEA region.
Corey Lanum - Americas Manager
Manages US office and relationships with customers
in US, Canada and APAC area.
Author of Visualizing Graph Data - quote ‘vgdweb’ for 39% off.
Purpose of Visualization
● To better understand the structure of the data
you are collecting
● To better understand the relationships
contained in the data you are collecting
Why Visualize Networks
● Some graph problems aren’t solved by visualization, for example recommendation engines
● Graph data is inherently visual
● Accessible by non-scientists
● Can convey a deeper understanding of the data
Who Uses Graph Visualization
Finance and Insurance● Fraud discovery and investigation
● Regulatory compliance
Information Technology
● Network topology
● Risk assessment
Government
● Defense and intelligence
● Law enforcement
Oil and Gas
● Physical infrastructure
Understanding Terror Networks
● In 2004, Marc Sageman assembled a series of tables of 176 known communications and meetings between Al-Qaeda members and sympathizers.
● The communications take place between individuals so can be treated as a matrix
● Communications can be assembled into an association matrix - showing who knows each other by making marks in a grid
● This can be binary, either there’s a link or not, or weighted, with values on those connections
We’re starting to get some value out of this data, but what about when we want to visualize it?
Understanding Terror Networks
● The problem with Social Network data is often too much data to visualize
● Data can be useful to identify influencers, groups, propagation throughout the network of information
● SNA centrality measures, Social Network Analysis can be particularly useful here.
● Twitter and public Facebook tweets/posts/likes can be downloaded via Netlytic, a free tool for SNA data. Commercial options are more robust.
Social Networks - Identify Influencers
Review Fraud
● User written reviews are critical to online commerce
● Sites like Amazon, Yelp, TripAdvisor all put their reviews front-and-center to drive sales and site visits
● One study showed a 19% increase in revenue for a 1 star increase in average rating on Yelp
● This creates an ‘unhealthy ecosystem’ of fraudsters looking to artificially inflate or deflate reviews of products
● The volume makes it difficult to read each review individually
● Graph visualization can help
● I have not done this for my book
Review Fraud
● First, we need to format the review data as a graph
● The nodes will be the concrete things in our data○ First, the products/businesses being
reviewed○ Second, the review itself, which has
the date/time of the review submission and the star rating as a property
○ Third, the known properties of the reviewer such as device fingerprint, IP address, and e-mail address
● The edges represent the links between the reviewer, the review, and the business
Review Fraud
● Let’s zoom in to identify suspicious patterns● On the left, we have used the KeyLines timebar to zoom in to
reviews only posted on a single day● On the right, we see multiple negative reviews of a restaurant that
day from users with no other activity ever. Is this legitimate or an attempt to defame?