Clustering social network profiles using possibilistic C-means algorithm

Abstract

Social networking has become one of the most useful tools in modern society. Unfortunately, terrorists are taking advantage of the easiness of accessing social networks and they have set up profiles to recruit, radicalize and raise funds. Most of these profiles have pages that existing as well as new recruits to join the terrorist groups see and share information. Therefore, there is a potential need of detecting terrorist communities in social network in order to search for key hints in posts that appear to promote the militants cause. Community detection has recently drawn intense research interest in diverse ways. However, it represents a big challenge of practical interest that has received a great deal of attention. Social network clustering allows the labeling of social network profiles that is considered as an important step in community detection process. In this paper, we used possibilistic c-means algorithm for clustering a set of profiles that share some criteria. The use of possibility theory version of k-means algorithm allows more flexibility when assigning a social network profile to clusters. We experimentally showed the efficiency of the use of possibilistic c-means algorithm through a detailed tweet extract, semantic processing and classification of the community detection process.