The tools of social network analysis – centrality measures, clustering, graph-traversal algorithms, community detection and so forth – are largely based on mathematical network theory. There is very little in these techniques that actually requires that the data represents social activity. This presentation will show how these techniques can be applied to data from areas such as geo, the Wikipedia link graph and linguistics.
We’ll show how to take tabular or textual data and derive graph representations from it that can be used to apply these techniques. We’ll discuss practical applications of these techniques in delivering new features for web applications. We’ll also show how the powerful visualisation tool Gephi can be used to explore the data once it’s in graph form.
This talk will be partly based on content from an Ignite talk given at Strata NYC 2011: http://slideshare.net/mattb/place-graphs-are-the-new-social-graphs
Matt Biddulph is an independent creative technologist. He was co-founder of Dopplr, the social network for smarter travel acquired by Nokia in 2009. He started out in 1994 building search engines on CD-ROM, and now specialises in digital media, social software and putting data on the web.
For information on exhibition and sponsorship opportunities at the conference, contact Susan Stewart at firstname.lastname@example.org.
For information on trade opportunities with O'Reilly conferences contact Kathy Yu at mediapartners
For media-related inquiries, contact Maureen Jennings at email@example.com
View a complete list of Strata contacts