A Comparative Analysis of Identifying Influential Users in Online Social Networks

摘要

It has been the one of the primary efforts of social studies for decades to find influential people in a society due to its numerous applications, such as spreading ideas and practices. In today's world, Online Social Networks (OSNs) support their users by analyzing the behaviors of their users and helping them to maintain/create social presence. Several techniques for analyzing social networks have been developed, to gauge quantitative properties or qualitative aspects. However, online social network analysis poses novel challenges both to computer and social scientists. Influential User is a unique web and social phenomenon affecting tastes and behaviors of their users and helping them to maintain/create friendships. Since the influence of good number of connections/friends needs to be evaluated for each user, it is difficult to infer precisely which among them is influential. Many models ranging from Associative Rule mining to analyzing a bunch of social network metrices have been proposed. In all of the proposed methods, the very first step is to quantify the amount of influence an individual exerts on another. There can be many different ways to calculate the power of user. We demonstrate how these models are able to find the influential users. Besides we do an analytical comparison between different methods. Eventually, we conclude that there is no single method that is better than others as the problem is hard, so the space for new algorithms is wide open.