Presentation	1999/3/12 Top-down Clustering Based on MDL Criterion for Large Document Set Kazunori MATSUMOTO, Keiko AOKI, Keiichirou HOASHI, Kazuo HASHIMOTO,
PDF Download Page	PDF download Page Link
Abstract(in Japanese)	(See Japanese page)
Abstract(in English)	Conventional top-down clustering algorithm uses a random sampling technique for the construction of section clusters. Authors proposes an evaluation function based on the entropy maximization principle which selects the best section cluster. Empirical result shows the retrieval accuracy of the proposed method is better than that of the conventional one. We also propose an evaluation function based on MDL criterion.
Keyword(in Japanese)	(See Japanese page)
Keyword(in English)	Similarity Based Text Retrieval / Clsuering / MDL criterion
Paper #	NLC98-56
Date of Issue

Conference Information
Committee	NLC
Conference Date	1999/3/12(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	Natural Language Understanding and Models of Communication (NLC)
Language	JPN
Title (in Japanese)	(See Japanese page)
Sub Title (in Japanese)	(See Japanese page)
Title (in English)	Top-down Clustering Based on MDL Criterion for Large Document Set
Sub Title (in English)
Keyword(1)	Similarity Based Text Retrieval
Keyword(2)	Clsuering
Keyword(3)	MDL criterion
1st Author's Name	Kazunori MATSUMOTO
1st Author's Affiliation	KDD R&D Laboratories INC.()
2nd Author's Name	Keiko AOKI
2nd Author's Affiliation	KDD R&D Laboratories INC.
3rd Author's Name	Keiichirou HOASHI
3rd Author's Affiliation	KDD R&D Laboratories INC.
4th Author's Name	Kazuo HASHIMOTO
4th Author's Affiliation	KDD R&D Laboratories INC.
Date	1999/3/12
Paper #	NLC98-56
Volume (vol)	vol.98
Number (no)	660
Page	pp.pp.-
#Pages	6
Date of Issue