Viewing Implicit Social Networks As Bipartite Graphs

In this paper, we describe a new model for thinking about implicit social networks, which are defined when users interact by modifying a common shared object. These are also called "knowledge sharing" networks and are in contrast to the typical person-to-person social networks. The properties of these networks are important to many data mining applications including recommendation systems, personalization, and other networks. The model consists of a bipartite graph where users and interaction points are two classes of nodes, connected by edges when the users interact with the shared objects represented by the interaction points. For example, an interaction point for an online book store might be the reviews for a certain book. In the end, we conclude that the information in the network structure of the bipartite graph contains significant useful information on the implicit social network and can be used as an effective data structure for analysis and visualization.