A simplicial complex, a hypergraph, structure in the latent semantic space of document clustering

Tsau Young Lin, I. Jen Chiang

研究成果: 雜誌貢獻文章

22 引文 (Scopus)

摘要

This paper presents a novel approach to document clustering based on some geometric structure in Combinatorial Topology. Given a set of documents, the set of associations among frequently co-occurring terms in documents forms naturally a simplicial complex. Our general thesis is each connected component of this simplicial complex represents a concept in the collection. Based on these concepts, documents can be clustered into meaningful classes. However, in this paper, we attack a softer notion, instead of connected components, we use maximal simplexes of highest dimension as representative of connected components, the concept so defined is called maximal primitive concepts. Experiments with three different data sets from Web pages and medical literature have shown that the proposed unsupervised clustering approach performs significantly better than traditional clustering algorithms, such as k-means, AutoClass and Hierarchical Clustering (HAG). This abstract geometric model seems have captured the latent semantic structure of documents.

原文英語
頁(從 - 到)55-80
頁數26
期刊International Journal of Approximate Reasoning
40
發行號1-2
DOIs
出版狀態已發佈 - 七月 2005

指紋

Document Clustering
Simplicial Complex
Hypergraph
Clustering algorithms
Websites
Semantics
Connected Components
Topology
Experiments
Unsupervised Clustering
Geometric Model
K-means Clustering
Hierarchical Clustering
Geometric Structure
Higher Dimensions
Clustering Algorithm
Attack
Concepts
Document clustering
Term

ASJC Scopus subject areas

  • Statistics and Probability
  • Electrical and Electronic Engineering
  • Statistics, Probability and Uncertainty
  • Information Systems and Management
  • Information Systems
  • Computer Science Applications
  • Artificial Intelligence

引用此文

A simplicial complex, a hypergraph, structure in the latent semantic space of document clustering. / Lin, Tsau Young; Chiang, I. Jen.

於: International Journal of Approximate Reasoning, 卷 40, 編號 1-2, 07.2005, p. 55-80.

研究成果: 雜誌貢獻文章

@article{ddf648c0321542129499a588a6a37107,
title = "A simplicial complex, a hypergraph, structure in the latent semantic space of document clustering",
abstract = "This paper presents a novel approach to document clustering based on some geometric structure in Combinatorial Topology. Given a set of documents, the set of associations among frequently co-occurring terms in documents forms naturally a simplicial complex. Our general thesis is each connected component of this simplicial complex represents a concept in the collection. Based on these concepts, documents can be clustered into meaningful classes. However, in this paper, we attack a softer notion, instead of connected components, we use maximal simplexes of highest dimension as representative of connected components, the concept so defined is called maximal primitive concepts. Experiments with three different data sets from Web pages and medical literature have shown that the proposed unsupervised clustering approach performs significantly better than traditional clustering algorithms, such as k-means, AutoClass and Hierarchical Clustering (HAG). This abstract geometric model seems have captured the latent semantic structure of documents.",
keywords = "Association rules, Document clustering, Hierarchical clustering, Simplicial complex, Topology",
author = "Lin, {Tsau Young} and Chiang, {I. Jen}",
year = "2005",
month = "7",
doi = "10.1016/j.ijar.2004.11.005",
language = "English",
volume = "40",
pages = "55--80",
journal = "International Journal of Approximate Reasoning",
issn = "0888-613X",
publisher = "Elsevier Inc.",
number = "1-2",

}

TY - JOUR

T1 - A simplicial complex, a hypergraph, structure in the latent semantic space of document clustering

AU - Lin, Tsau Young

AU - Chiang, I. Jen

PY - 2005/7

Y1 - 2005/7

N2 - This paper presents a novel approach to document clustering based on some geometric structure in Combinatorial Topology. Given a set of documents, the set of associations among frequently co-occurring terms in documents forms naturally a simplicial complex. Our general thesis is each connected component of this simplicial complex represents a concept in the collection. Based on these concepts, documents can be clustered into meaningful classes. However, in this paper, we attack a softer notion, instead of connected components, we use maximal simplexes of highest dimension as representative of connected components, the concept so defined is called maximal primitive concepts. Experiments with three different data sets from Web pages and medical literature have shown that the proposed unsupervised clustering approach performs significantly better than traditional clustering algorithms, such as k-means, AutoClass and Hierarchical Clustering (HAG). This abstract geometric model seems have captured the latent semantic structure of documents.

AB - This paper presents a novel approach to document clustering based on some geometric structure in Combinatorial Topology. Given a set of documents, the set of associations among frequently co-occurring terms in documents forms naturally a simplicial complex. Our general thesis is each connected component of this simplicial complex represents a concept in the collection. Based on these concepts, documents can be clustered into meaningful classes. However, in this paper, we attack a softer notion, instead of connected components, we use maximal simplexes of highest dimension as representative of connected components, the concept so defined is called maximal primitive concepts. Experiments with three different data sets from Web pages and medical literature have shown that the proposed unsupervised clustering approach performs significantly better than traditional clustering algorithms, such as k-means, AutoClass and Hierarchical Clustering (HAG). This abstract geometric model seems have captured the latent semantic structure of documents.

KW - Association rules

KW - Document clustering

KW - Hierarchical clustering

KW - Simplicial complex

KW - Topology

UR - http://www.scopus.com/inward/record.url?scp=19044395829&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=19044395829&partnerID=8YFLogxK

U2 - 10.1016/j.ijar.2004.11.005

DO - 10.1016/j.ijar.2004.11.005

M3 - Article

AN - SCOPUS:19044395829

VL - 40

SP - 55

EP - 80

JO - International Journal of Approximate Reasoning

JF - International Journal of Approximate Reasoning

SN - 0888-613X

IS - 1-2

ER -