Community Detection Algorithms: A Critical Review

Community Detection Algorithms: A Critical Review

Akib Mohi Ud Din Khanday, Syed Tanzeel Rabani, Qamar Rayees Khan, Fayaz Ahmad Khan
Copyright: © 2023 |Pages: 17
DOI: 10.4018/978-1-6684-6909-5.ch004
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Modern networks, like social networks, can typically be characterised as a graph structure, and graph theory approaches can be used to address issues like link prediction, community recognition in social network analysis, and social network mining. The community structure or cluster, which is the organisation of vertices with numerous links joining vertices of the same cluster and comparatively few edges joining vertices of different clusters, is one of the most important characteristics of graphs describing real systems. People post content on social media platforms and others comment, share, and like their messages. There are various approaches in finding the communities on online social networks. In this chapter an overview of community structure is provided. A critical analysis is being done on various community detection algorithms.
Chapter Preview
Top

Literature Review

Humans have the nature to make communities in the real world, which is reflected in social media. Newman and Girvan (2004) proposed algorithm for detecting group structure. The two major characteristics of the algorithm are: The first step is to remove edges from the network iteratively, forming communities from the networks; the second step is to recalculate the edges after each removal. The algorithms used are more effective at detecting group structure in both machine and real-world network data. Finding a faster version of the algorithm since this algorithm becomes intractable for larger systems. The algorithm has been improved to reduce the computational complexity.

Complete Chapter List

Search this Book:
Reset