Frequent subpatterns distribution in social network analysis

Abstract

Discovering the frequencies of a small subgraph in social networks graph is the most challenging part. Because social networking graphs have very few structural constraints which appear mostly in frequent patterns only. Many graph-based mining algorithms are available to solve this problem like Frequent Subgraph Mining and Motif Detection. Both algorithms find a small set of subgraph that repeats themselves in a large network for a specific amount of time. But motif detection gives the deep understanding of network functionality and also uncovers the structural property of complex network. The problem with analyzing the frequent patterns is their detection comes under computationally challenging problems. We can solve this problem by using sampling method which is not bound to the size of the graph and can compute millions of nodes. The full distribution of frequent pattern gives the statistical information about the underlying network which helps to understand network better and also use in analysis of the network. By using pattern distribution, we can uncover and identify the different aspect of Social Network which can be used in many real-life applications.