Data Mining on Social Interaction Networks

Social media and social networks have already woven themselves into the veryfabric of everyday life. This results in a dramatic increase of social datacapturing various relations between the users and their associated artifacts,both in online networks and the real world using ubiquitous devices. In thiswork, we consider social interaction networks from a data mining perspective -also with a special focus on real-world face-to-face contact networks: Wecombine data mining and social network analysis techniques for examining thenetworks in order to improve our understanding of the data, the modeledbehavior, and its underlying emergent processes. Furthermore, we adapt, extendand apply known predictive data mining algorithms on social interactionnetworks. Additionally, we present novel methods for descriptive data miningfor uncovering and extracting relations and patterns for hypothesis generationand exploration, in order to provide characteristic information about the dataand networks. The presented approaches and methods aim at extracting valuableknowledge for enhancing the understanding of the respective data, and forsupporting the users of the respective systems. We consider data from severalsocial systems, like the social bookmarking system BibSonomy, the socialresource sharing system flickr, and ubiquitous social systems: Specifically, wefocus on data from the social conference guidance system Conferator and thesocial group interaction system MyGroup. This work first gives a shortintroduction into social interaction networks, before we describe severalanalysis results in the context of online social networks and real-worldface-to-face contact networks. Next, we present predictive data mining methods,i.e., for localization, recommendation and link prediction. After that, wepresent novel descriptive data mining methods for mining communities andpatterns