Detecting communities in social networks using unnormalized spectral clustering incorporated with Bisecting K-means

Social network analysis has gained much attention now a days. Social networks can be represented as a graph. In Social network analysis each individual is represented as a node and the relationship between them is represented as an edge in the graph. In social networks, community structure indicates that nodes within the group are densely connected and the connections between groups are weak. We propose a novel method called unnormalized spectral clustering incorporated with Bisecting K-means to identify communities in social networks. Our method is able to identify strongly connected groups of nodes with weaker connections between groups. We performed the experiments on three real-world network datasets, namely Zachary karate club dataset, American College Football dataset and Bottlenose Dolphin Network dataset and find that it performs better than the benchmark algorithm unnormalized spectral clustering with K-means. Experimental results on real world network datasets show the effectiveness of our method.