Presentation 2006-07-14
Parameter Setting for a Clustering Method through an Analytical Study of Real Data
Sophoin KHY, Yoshiharu ISHIKAWA, Hiroyuki KITAGAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In our previous papers, we have proposed a novelty-based document clustering method. The clustering method generates clusters reflecting current trend of recent topics. However, the problem with this clustering method is the setting of the exponential decaying factor parameter through which a document loses its value as time passes. In the clustering method, the parameter value is assumed a user-specified parameter and only one value is used for all types of documents regardless of which categories they are in. In this paper, we will show an analytical study on the real dataset. The objective of this study is to find the relationship between the decaying behavior of each document with their corresponding category types which will be helpful in the setting of parameter values in the novelty-based document clustering method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) clustering / parameter setting / forgetting factor / exponential decay
Paper # DE2006-99
Date of Issue

Conference Information
Committee DE
Conference Date 2006/7/7(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 ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Parameter Setting for a Clustering Method through an Analytical Study of Real Data
Sub Title (in English)
Keyword(1) clustering
Keyword(2) parameter setting
Keyword(3) forgetting factor
Keyword(4) exponential decay
1st Author's Name Sophoin KHY
1st Author's Affiliation Department of Computer Science, Graduate School of Systems and Information Engineering, University of Tsukuba()
2nd Author's Name Yoshiharu ISHIKAWA
2nd Author's Affiliation Information Technology Center, Nagoya University
3rd Author's Name Hiroyuki KITAGAWA
3rd Author's Affiliation Department of Computer Science, Graduate School of Systems and Information Engineering, University of Tsukuba:Center for Computational Sciences, University of Tsukuba
Date 2006-07-14
Paper # DE2006-99
Volume (vol) vol.106
Number (no) 150
Page pp.pp.-
#Pages 6
Date of Issue