DISCOVERY OF GATEKEEPERS ON INFORMATION DIFFUSION FLOWS USING PROCESS MINING

Online social network services (SNS) such as Twitter and Facebook are currently representative means of disseminating information on the Web. It therefore is crucial to internet marketing to understand the dynamics of information diffusion in the online social networks. To this end, the problem of discovering information diffusion processes is dealt with in this paper. A probabilistic approach to process discovery is presented based on an extended hidden Markov model, considering the log data which are extracted from the online social network services. Specifically, the users are first grouped from the SNS log data based on their interactions using a few clustering algorithms. The process discovery algorithm which is extended from the hidden Markov model is applied to the user clusters to reflect on the probabilistic dissemination among the user clusters. The proposed method is illustrated with a real SNS data which was gathered from a Facebook fan page. It is expected that the method can help to comprehend the information dynamics in online social networks by visualizing probabilistic information diffusion process according to the user group