Generating hypergraph of term associations for automatic document concept clustering

I. J. Chiang, Tsau Young Lin, J. Y J Hsu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper presents a novel approach to document clustering using hypergraph decomposition. Given a set of documents, the associations among frequently co-occurring terms in any of the documents define naturally a hypergraph, which can then be decomposed into connected components at various levels. Each connected component represents a primitive concept in the collection. The documents can then be clustered based on the primitive concepts. Experiments with three different data sets from web pages and medical literatures have shown that the proposed unsupervised clustering approach performs significantly better than traditional clustering algorithms, such as k-means, AutoClass and Hierarchical Clustering (HAC). The results indicate that hypergraphs are a perfect model to capture association rules in text and is very useful for automatic document clustering.

Original languageEnglish
Title of host publicationProceedings of the Eighth IASTED International Conference on Artificial Intelligence and Soft Computing
EditorsA.P. Pobil
Pages181-186
Number of pages6
Publication statusPublished - 2004
EventProceedings of the Eighth IASTED International Conference on Atificial Intelligence and Soft Computing - Marbella, Spain
Duration: Sep 1 2004Sep 3 2004

Other

OtherProceedings of the Eighth IASTED International Conference on Atificial Intelligence and Soft Computing
CountrySpain
CityMarbella
Period9/1/049/3/04

    Fingerprint

Keywords

  • Association Rules
  • Concept
  • Connected Components
  • Decomposition
  • Document Clustering
  • Hypergraph

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Chiang, I. J., Lin, T. Y., & Hsu, J. Y. J. (2004). Generating hypergraph of term associations for automatic document concept clustering. In A. P. Pobil (Ed.), Proceedings of the Eighth IASTED International Conference on Artificial Intelligence and Soft Computing (pp. 181-186)