Overlapping community detection algorithms in complex networks based on the fuzzy spectral clustering

Aim at the problem that the most of algorithm for community discovery assume that community don´t overlap with each other, the paper combined with spectrum graph theory and fuzzy sets theory to analyze the community structures in complex networks, the paper proposes a fuzzy spectral clustering algorithm FSC for discovery overlapping community. The basic idea of FSC is to describe communities membership of network nodes with membership degree function, the community of each node belong to the community by the membership. By two different types of real network simulation, the results demonstrate the feasibility and effectiveness of the approach.