COMMUNITY DETECTION IN SOCIAL NETWORK USING GRAPH CLUSTERING METHODS

Abstract

Community detection is a major analysis area in social media analysis where we identify the social
network construction. Detecting communities is of main interest in sociology, computer science, biology, and
methods, where systems are usually described as graphs. Community detection aims at discovery clusters as
subgraphs within a specified network. A community is then a cluster where many edges link nodes of the same
group and few edges link nodes of different clusters. With the democratization of the internet, communicating
and sharing information is more manageable than ever. Community detection is a solution to understanding the
structure of complex networks and finally extracting useful information from them. In Facebook, the existence
of communities (groups) is a critical question; thus, many researchers focus on potential communities by using
techniques like data mining and web mining. In this paper, the community detection for a Facebook social
network is presenting.Ithas developed in network science to find groups within complex systems depicted on a
graph.The proposed model divided into various phases such as a) sub-graph discovery, b) vertex clustering, c)
community quality optimization, d) divisive, and e) model-based. For community detection defining the
consistency between social particles, Social media applied Social Balance, Social status theory, Social
correlation theory, and finally applying K-means clustering over facebook data set. The Experimental conducted
on Face book social media dataset with multiple edges and nodes. The experimental results shows that the
proposed model gets higher accuracy in community detection compared with state of the art methods.