Overlapping community detection in complex networks using fuzzy theory, balanced link density, and label propagation

Abstract

Complex networks represent various real-world systems. Overlapping community detection is one of the critical tasks in studying these networks and has significance to a wide variety of applications, including the exploration of online social networks because of the natural attitude of persons to participate in multiple communities at the same time. Despite a large number of existing community detection algorithms for detecting disjoint communities, the efficient and fast uncovering of overlapping communities has remained a challenging problem. To provide an efficient solution, on the one hand, the balanced link density label propagation (BLDLP) algorithm, proposed by the authors of the current study, is a fast, stable, and efficient method for disjoint community detection. On the other hand, the fuzzy theory is a worthwhile approach for overlapping community detection since it provides the membership rate of the overlapping nodes as well as the detection of overlapping communities. Hence, in this paper, based on the synergy of the BLDLP algorithm and the fuzzy theory, a novel method, called fuzzy BLDLP, for overlapping community detection is proposed.Fuzzy BLDLP is fast and efficient. The proposed method needs no prior information about the number of network communities to discover them. The experiments on both synthetic and real-world known networks, including Zachary, Dolphins, and COVID-19 Co-authorship, have revealed that the proposed method successfully detects the overlapping nodes and communities and hence is comparable with the state-of-the-art overlapping community detection algorithms in terms of recall, precision, F-score and overlapping normalized mutual information.

An effective overlapping community merging method oriented to multidimensional attribute social networks

“…Yang et al (2017) proposed the LFM base on Jaccard (LFMJ) selecting seed nodes through random walk method and node's degree for avoiding the disadvantages of randomly selecting seed nodes. In order to solve the problem of how to find overlapping communities efficiently and quickly, Jokar et al (2022) proposed the fuzzy BLDLP method, which does not require the prior information of the number of communities, and improves the effectiveness of community detection. Shang et al (2022) expanded the community according to the motif degree and modular function, taking into account the higher‐order information of the network.…”

Community detection from fuzzy and higher-order perspectives

“…fuzzy BLDLP [30]) usually iteratively propagates and updates fuzzy memberships of nodes according to the edge connection intensity with neighbor nodes. Such methods are usually computationally efficient, but have the problems of poor stability and error amplification, due to the randomness in label propagation process.…”

Presenting a Novel Method for Identifying Communities in Social Networks Based on the Clustering Coefficient

“…Many community identification algorithms have been presented for analyzing social networks. Identifying communities can be considered an optimization problem; from this point of view, several methods based on the maximization of the famous modularity criterion have been presented for identifying communities [18]. Among the famous methods in this field, we can mention Newman's greedy algorithm (FN) [19].…”

“…The proposed CC-BF algorithm appears to be designed to address these gaps and is tested on various types of network data. [16] used to make a comparison in (S27, S45) more than 5000 S27, s28,s45,s51 4 A rapid method for modularity [18] used to make a comparison in (S13) 2000 S,10, S13, S18 3 MapInfo algorithm [19] used as a comparative tool more than 3500 S32, S36 2 Algorithm for maximizing expectations [23] used as a comparative tool more than 6000 S16, S24, S29, S41, S47 5 Community detection algorithm for binary graphs [24] newly suggested method 1500 S03 1 Algorithm for eigenvector label propagation [29] newly suggested method more than 5000 S40, S46, S49, S53 4 Spectral clustering-based technique for left-to-right oscillation [30] newly suggested method 5000 S34, S45 2 Algorithm for evaluation and identification in the community [31] newly suggested method more than 4000 S22, S05 2 Eigenvector algorithm in front…”