Indian Journal of Science and Technology

Abstract

Background: Social Network Analysis is a prominent field of research that captures the attention of numerous data mining expert. Social networks are visualized as network graphs, and identifying communities involves the identification of densely connected nodes. The exploration of community detection in online social networks is an essential field of research. This review paper presents an extensive examination of the latest methodologies and approaches utilized for isolated and overlapped community detection specifically in online social networks. Objective: To provide a comprehensive overview of the existing literature on community detection techniques for social networks with isolated and overlapped communities. The review aims to identify the key challenges associated with community detection in such networks and to review the various algorithms and methods that have been proposed to address these challenges. Additionally, this review intends to compare the performance of different community detection techniques on networks with isolated and overlapped communities, and to highlight their strengths and weaknesses. Ultimately, the goal of the review is to provide researchers and practitioners with a better understanding of the current state of the art in community detection for social networks, and to help guide future research in this important area. Methods: A comprehensive literature search was conducted on quality databases viz. Scopus, Web of Science, IEEE Xplore, and Science Direct using relevant keywords, and the selected articles were screened for inclusion/ exclusion based on their titles, abstracts, and keywords. The following are some search keywords that can be used for searching community detection papers: Community detection, Graph clustering, Network partitioning, Modularity optimization, Community structure, Graph-based clustering, Network community detection, Overlapping communities, Community detection algorithms, and Evaluation metrics for community detection. Comparative analysis was performed on the chosen articles, considering factors such as algorithms, methodologies, datasets, and evaluation metrics. The results were presented using comparative tables to present the findings. The methodology ensured a comprehensive review of recent literature on community detection, providing valuable insights and trends in the field. Findings: The review identified various algorithms and methods that have been proposed to address these challenges, including modularity optimization, spectral clustering, label propagation, and network embeddings. The paper also found that many of these algorithms can be adapted to handle networks with isolated and overlapped communities, but there is no single algorithm that is universally effective. The review highlighted that the performance of community detection techniques can be influenced by several factors, such as network size, density, and community structure. The paper found that some algorithms perform better than others in certain scenarios and that a combination of methods may be necessary to achieve optimal results. The review found that there are several challenges associated with community detection in such networks, including the presence of noise, sparsity, and overlapping communities. Novelty: This paper presents several novel contributions to the field of community detection in social networks. One key novelty of the paper is its comprehensive review of the existing literature on community detection techniques for networks with isolated and overlapped communities, which provides a valuable resource for researchers and practitioners working in this area. Another novelty of the paper is its identification of the key challenges associated with community detection in such networks, and its discussion of the various algorithms and methods that have been proposed to address these challenges. The paper also presents a comparison of the performance of different community detection techniques on networks with isolated and overlapped communities, which can help guide future research in this area. Finally, the paper proposes several new approaches for community detection in social networks with complex community structures, which can help advance our understanding of social networks and their potential as powerful tools for communication, collaboration, and information sharing. Overall, the paper’s novel contributions make a significant contribution to the field of community detection in social networks.

Keywords: Social network analysis; community detection; influential user; overlapping communities; dynamic networks