the increasing popularity of social networking sites drives its users towards social behaviors and social conventions in the context of social networking. Recommendation or forecasting about the users of social networks is a new research area which is gaining researchers’ attention. Communities in social networks are the common platform where users have common interest or thoughts. Users share their thoughts, news and views in social communities by posting on the wall, commenting or liking others post, inviting friends to that community etc. In this paper, we proposed a novel mining rule based community recommendation approach for social networking sites. The experimental result for this approach gives 94% accuracy with model dataset.