Abstract
Community detection is a key aspect for understanding network structures and uncovers the underlying functions or characteristics of complex systems. A community usually refers to a set of nodes that are densely connected among themselves, but sparsely connected to the remaining nodes of the network. Detecting communities has been proved to be a NP-hard problem. Therefore, evolutionary based optimization approaches can be used to solve it. But a primary challenge for them is the higher computational complexity when dealing with large scale networks. In this respect, a COMpression based Multi-Objective Evolutionary Algorithm with Decomposition (Com-MOEA/D) for community detection is proposed where the network is first compressed to a much more smaller scale by exploring network topologies. After that, a framework of multi-objective evolutionary algorithm based on decomposition is applied, in which a local information based genetic operator is proposed to speed up the convergence and improve the accuracy of the Com-MOEA/D algorithm. Experimental results on both real world and synthetic networks show the superiority of the proposed method over several state-of-the-art community detection algorithms.
Highlights
Many real-world systems such as information, biological, transportation systems and social networks can be modelled as networks, which appropriately describe system elements and the relationships between them
The multi-objective evolutionary algorithms (MOEAs) have demonstrated their competitive performance in coping with the community detection problem, and in this paper, we adopt the general framework of decomposition based MOEA (MOEA/D) and propose a network compression based multi-objective evolutionary algorithm, termed Com-MOEA/D, to further improve the quality of detected communities as well as the efficiency
EXPERIMENTAL DESIGN 1) COMPARISON ALGORITHMS In the experiments, the comparison algorithms are among the Com-MOEA/D and popular EA based multi-objective optimization algorithms, as well as classical ones
Summary
Many real-world systems such as information, biological, transportation systems and social networks can be modelled as networks, which appropriately describe system elements and the relationships between them. Several other decomposition based MOEAs, named LMOEA [22], MODBSA/D [23] and RMOEA [24], have been developed They all adopted the general framework of decomposition based multi-objective evolutionary algorithms to detect communities, and experimental results showed their effectiveness and efficiency. The MOEAs have demonstrated their competitive performance in coping with the community detection problem, and in this paper, we adopt the general framework of decomposition based MOEA (MOEA/D) and propose a network compression based multi-objective evolutionary algorithm, termed Com-MOEA/D, to further improve the quality of detected communities as well as the efficiency. Based on the compressed network, a multi-objective evolutionary algorithm based on decomposition is proposed to solve community detection problems It adopts the same framework as MOEA/D, and has two contradictory sub-objectives.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
