Bipartite graph partitioning and data clustering

Technical Report · Mon May 07 04:00:00 EDT 2001

DOI:https://doi.org/10.2172/816202· OSTI ID:816202

Zha, Hongyuan ^[1]; He, Xiaofeng; Ding, Chris; Gu, Ming; Simon, Horst D

---

1. LBNL Library

Many data types arising from data mining applications can be modeled as bipartite graphs, examples include terms and documents in a text corpus, customers and purchasing items in market basket analysis and reviewers and movies in a movie recommender system. In this paper, the authors propose a new data clustering method based on partitioning the underlying biopartite graph. The partition is constructed by minimizing a normalized sum of edge weights between unmatched pairs of vertices of the bipartite graph. They show that an approximate solution to the minimization problem can be obtained by computing a partial singular value decomposition (SVD) of the associated edge weight matrix of the bipartite graph. They point out the connection of their clustering algorithm to correspondence analysis used in multivariate analysis. They also briefly discuss the issue of assigning data objects to multiple clusters. In the experimental results, they apply their clustering algorithm to the problem of document clustering to illustrate its effectiveness and efficiency.

• Research Organization: Ernest Orlando Lawrence Berkeley National Laboratory, Berkeley, CA (US)
• Sponsoring Organization: USDOE Director, Office of Science. Office of Advanced Scientific Computing Research. Mathematical, Information, and Computational Sciences Division (US)
• DOE Contract Number: AC03-76SF00098
• OSTI ID: 816202
• Report Number(s): LBNL--47970
• Country of Publication: United States
• Language: English

Related Subjects

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE WASTES FROM NUCLEAR FACILITIES
99 GENERAL AND MISCELLANEOUS
ALGORITHMS
EFFICIENCY
MARKET
MINIMIZATION
MINING
MULTIVARIATE ANALYSIS