Influential node detection in social network during community detection

The proliferation of social media has inundated a gamut of research purviews e.g. Node classification, Community analysis, Behavioral psychological in social media. Out of these research issues the study of role of individual in terms of influential phrase plays a crucial role in deciding the future and popularity of any online social community. With the deep analysis of the community, the various parameters like: density, clustering coefficient, degree centrality, closeness centrality, eigenvector centrality etc. can be extracted with an effective and reliable way. These features can be easily exploited to determine the most influential individual /node in an online community. Motivated by the complexity of research problem of finding most influential nodes in an online evolving community, this paper is an effort to understand and model the complexity of the dynamics of different metrics of social network. The proposed scheme is corroborated with the rigorous result analysis.

I. Introduction

The analyses of social dynamics which includes the narrative of those distinctive features that characterize online social networks are acquiring a mounting importance in the modern era. Complex network into dense sub network called community. Community is one starts from a genuine position and is built in the order of a common interest. Community allows the nodes (i.e. people) to interact with each other and to develop relation with each other. Only those people who have keen interest in particular community would like to be a part of the community. Thus a community should be Authentic, Accessible and Adaptable. It is an abstruse problem that how the evolution of an online social community can be predicted. For detecting the evolution, first the network is needed to be partitioned into dense areas, known as communities.