Abstract
A social network enables individuals to communicate with each other by posting information, comments, messages, images, etc. In most applications, a social network is modelled by a graph with vertices and edges. Vertices represent individuals and edges represent social interactions between the individuals. A social network is said to have community structure if the nodes of the network can be grouped into sets of nodes such that each set is densely connected internally. The investigation of the community structure in the social network is an important issue in many domains and disciplines such as marketing and bio-informatics. Community detection in social networks can be considered as a graph clustering problem where each community corresponds to a cluster in the graph. The goal of conventional community detection methods is to partition a graph such that every node belongs to exactly one cluster. However, in many social networks, nodes participate in multiple communities. Therefore, a node’s communities can be interpreted as its social circles. Thus, it is likely that a node belongs to multiple communities. We propose in this paper a new overlapping community detection method which can be adopted for several real world social networks requiring non-disjoint community detection.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have