A CPN-based information propagation model in Online Social Networks

In Online Social Networks (OSNs), information is propagated by interaction among users. A user interacts with others through a variety of behaviors, which can promote the flow of information resource. The previous researches on information propagation lack the analysis of user behavior characteristics and the intuition of the dynamics of information propagation. A OSN system is a distributed concurrent system which is suitable to be modeled graphically with Petri net theory. Taking the Weibo platform as an example, this paper models and simulates its information propagation process using Colored Petri net (CPN). We adopt a modular modeling method. First, based on seven typical user behaviors, namely browsing (no action), posting, reposting, liking, commenting, replying and deleting, we model the information propagation process, which describes the spread of a microblog posted by the initiator of a topic. Then we count the number of actions of each type generated by users during the propagation process, which can evaluate the users' participation for the topic, that is, the popularity of the topic. Finally, we combine the two modules to obtain a complete information propagation model using sharing composition. In addition, we run the model in CPN Tools to simulate the spread of a microblog posted by the initiator of a topic. Moreover, we analyze and verify the state changes and user behavior sequences of the information propagation system using State Space Tools. Our work reveals that it is feasible and advantageous to apply Petri Net theory to social network analysis, so as to discover more information propagation rules in the future.