Bo Long, Zhongfei Zhang, Philip S. Yu, Tianbing Xue
Complex graphs, in which multi-type nodes are linked to each other, frequently arise in many important applications, such as Web mining, information retrieval, bioinformatics, and epidemiology. In this study, We propose a general framework for clustering on complex graphs. Under this framework, we derive a family of clustering algorithms including both hard and soft versions, which are capable of learning cluster patterns from complex graphs with various structures and statistical properties. We also establish the connections between the proposed framework and the traditional graph partitioning approaches. The experimental evaluation provides encouraging results to validate the proposed framework and algorithms.
Subjects: 12. Machine Learning and Discovery; 1.10 Information Retrieval
Submitted: Apr 10, 2008