Community Detection from Signed Social Networks Using a Multi-objective Evolutionary Algorithm

Abstract

In this paper, we propose a method for detecting communities from signed social networks with both positive and negative weights by modeling the problem as a multi-objective problem. In the experiments, both real world and synthetic signed networks whose size ranges from 100 to 1200 nodes are used to validate the performance of the new algorithm. A comparison is also made between the new algorithm and an effective existing algorithm, namely FEC. The experimental results show that our algorithm obtains a good performance on both real world and synthetic data, and outperforms FEC clearly.