Community Detection and Probability Estimation of Community Connectedness

Abstract

We present a methodology to analyze large social data sets based on a new community detection algorithm and based on the communities we find a method to detect the probability of a node to be part of a community. Our main aim is to find the communities based on locally computed score and later fit the scores in a distribution to find the probability of its connectedness in a community. Our work is mostly based on FOCS algorithm. Therefore in this article we refer to the FOCS algorithm often, describe it and then point out the changes brought by us.