[This article belongs to Volume - 54, Issue - 05]
Gongcheng Kexue Yu Jishu/Advanced Engineering Science
Journal ID : AES-21-07-2022-244

Title : Diversified Overlapping Community Detection Methods in Social Networks: A Survey
Monika, Dr. Veenu Mangat,

Abstract : The growing era of social networks and ubiquity of the Internet have paved the way for extensive research in the area of social network analysis. Wide range of research challenges have been identified in the domain of social network analysis such as how individuals interact with each other, and what is the structure of the social network when diverse interactions are taking place. Detection of overlapping communities is one of the upcoming research challenges due to the participation of individuals in diverse network groups at the same time. The main objective of this paper is to present a comprehensive review of overlapping community detection methods for various types of social networks. The paper provides a systematic segregation of overlapping community detection methods and draws inferences regarding their application. Since the main issues in community detection in social networks are- scalability considering the huge scale of networks, dynamic nature of associations, and time-variant participation in communities, this paper brings forth critical analysis of computational complexity of overlapping community detection methods in literature. The parameters that can be used for evaluation of these methods as well as listing of real-world and synthetic benchmark datasets used for evaluation of each clique and non-clique method have also been detailed. This paper is intended to serve as a ready reference for the researchers in the area of social network analysis to develop efficient and accurate community detection methods for current social networks.