Presentation 2007-07-03
Dimensionality Reduction via Latent Dirichlet Allocation for Document Clustering
Tomonari MASADA, Senya KIYASU, Sueharu MIYAHARA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we employ the latent Dirichlet allocation as a method for the dimensionality reduction of feature vectors and reveal its effectiveness in document clustering. In the evaluation experiment, we perform clustering on the document sets of Japanese and Korean Web news articles. We regard the categories assigned to each article as the ground truth of clustering evaluation. We compare the clustering results obtained by using the feature vectors whose entries are term frequencies with the results obtained by using the feature vectors whose dimensions are reduced by the latent Dirichlet allocation.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) document clustering / dimensionality reduction / latent Dirichlet allocation
Paper # DE2007-85
Date of Issue

Conference Information
Committee DE
Conference Date 2007/6/25(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Data Engineering (DE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Dimensionality Reduction via Latent Dirichlet Allocation for Document Clustering
Sub Title (in English)
Keyword(1) document clustering
Keyword(2) dimensionality reduction
Keyword(3) latent Dirichlet allocation
1st Author's Name Tomonari MASADA
1st Author's Affiliation Faculty of Engineering, Nagasaki University()
2nd Author's Name Senya KIYASU
2nd Author's Affiliation Faculty of Engineering, Nagasaki University
3rd Author's Name Sueharu MIYAHARA
3rd Author's Affiliation Faculty of Engineering, Nagasaki University
Date 2007-07-03
Paper # DE2007-85
Volume (vol) vol.107
Number (no) 131
Page pp.pp.-
#Pages 6
Date of Issue