Loading [MathJax]/extensions/MathZoom.js
An Analysis of the Dynamic Community Detection Algorithms in Complex Networks | IEEE Conference Publication | IEEE Xplore

An Analysis of the Dynamic Community Detection Algorithms in Complex Networks


Abstract:

Uncovering the dynamics of community structures in complex networks helps us to explore how such community structures change over time. But, understanding these structure...Show More

Abstract:

Uncovering the dynamics of community structures in complex networks helps us to explore how such community structures change over time. But, understanding these structures is very challenging, especifically in dynamic complex networks where network structure changes frequently and interaction between the individuals changes over time. Recently, many dynamic community detection algorithms have been introduced to capture the dynamics of network community structures. In this paper, we present a detailed analysis of the dynamic community detection algorithms in terms of computation time and accuracy. To provide detailed and extensive analysis, we tested dynamic algorithms on small, medium and large real-world network dataset. Based on the analysis results and network properties, we provide some guidelines that may help to choose the best dynamic community detection algorithms for the given dynamic complex networks.
Date of Conference: 26-28 February 2020
Date Added to IEEE Xplore: 16 April 2020
ISBN Information:

ISSN Information:

Conference Location: Buenos Aires, Argentina

I. Introduction

Today's real-world networks such as social networks (e.g., Twitter, Facebook, etc.) and collaboration networks are complex and dynamic in nature [1]. Uncovering the dynamics of complex networks (i.e., the networks nodes or edges change over time) is a core research topics over the past few years [2]. Community structure (i.e., group of nodes that are strongly connected internally, but sparsely connected externally) is an important characteristic of the dynamic complex networks [3], [4]. For instance, communities of social networks could be organizational units; communities in the collaboration networks might represent a group of people working together on similar topics. Identifying communities in dynamic networks is a very challenging task. There are several methods that have been introduced for community finding in the last one and half decade, such as modularity optimization [5], clique percolation [6], greedy algorithms [7], label propagation [8], random-walk based [9], and topology based [3] methods. Nearly all these methods are developed for static networks.

References

References is not available for this document.