Community detection in social network based on useras social activities

Abstract

Listen

Translate article

A social networking provides a interface where users can share, connect, interact and build social relations with other users. Communities in a social network are virtual groups where people with common interests and activities like to communicate. Increasing number of social network and users has proportionally increased demand of communities. Profile attribute based communities have a major drawback as they consider only similarity of attribute values among community user not on their interest and social activities. Hence, a new approach to detect community on the basis of common interest, social activities and interaction among users is being proposed. In this approach a new parameter Common Social Activity (CSA) is derived and used along with other activity parameter to find community for a selected seed user. Activity parameters are given as input to k-means algorithm, which results to groups of user referred as community. These communities reflect groups of users having active social relationship and common interest which is also helpful for updates and marketing of new products and technology.