L1-Graph Based Community Detection in Online Social Networks
出版社：
摘要内容：
Detecting community structures in online social network isa challenging job for traditional algorithms, such as spectral cluster-ing algorithms, due to the unprecedented large scale of the network. Inthis paper, we present an e?cient algorithm for community detectionin online social network, which chooses relatively small sample matrixto alleviate the computational cost. We use l-graph to construct thesimilarity graph and integrate the graph laplacian with random walk indirected social network. The experimental results show the e?ectivenessof the proposed method.
关键词：
- l-graph, Spectral clustering algorithm, Graph Laplacian,Laplacian regularizer
