Talkative Friend Algorithm for inferring ties in Social Networks

Abstract

In recent times there has been an exponential growth in Social networks. Virtual communication has become the order of the day with the Online social Networks catering to this need. The huge amount of activity occurring on social media and the data that has been generated on social networks has kindled the interest of researchers to work on the Social Network Analytics (SNA). There are a variety of problems addressed in SNA, Link Prediction is one among them. The problem of Link Prediction is to infer a connection in a social network which could occur at a later point in time. This paper proposes an algorithm to infer future ties on social networks based on nodes which communicate more frequently with each other. The algorithm was tested on Data taken from real world networks and it has outperformed the existing baseline methods.