Author : Sanjiv Sharma and G.N. Purohit

Many real-world social networks are intimately organized according to a community structure. Most social networks are dynamic and connections between people change naturally overtime. Researchers have begun to consider the problem of tracking the formation of groups of users in social network. The situation is complicated by the fact that subgroups may split or merge, so that cohesiveness is not necessarily a property of a single subgroup, but may sometimes relate to a family of one or more related subgroups. However, in general, cohesive families of subgroups at one time period should be similar to corresponding subgroups at a different time period. Similarity is a topic that has received attention in a wide variety of scientific fields and a number of approaches are available for the measurement of similarity. This paper describes an efficient way for finding similarity in subgroups or clusters and tracking community which persist over time in dynamic networks.

Keywords: social network, community, similarity, groups

Article published in International Journal of Current  Engineering  and Technology, Vol.3,No.1 (March- 2013)