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.
|主出版物標題||Proceedings of the Eighth IASTED International Conference on Artificial Intelligence and Soft Computing|
|出版狀態||已發佈 - 2004|
|事件||Proceedings of the Eighth IASTED International Conference on Atificial Intelligence and Soft Computing - Marbella, 西班牙|
持續時間: 九月 1 2004 → 九月 3 2004
|其他||Proceedings of the Eighth IASTED International Conference on Atificial Intelligence and Soft Computing|
|期間||9/1/04 → 9/3/04|
ASJC Scopus subject areas