Presentation 2017-10-13
Multiple Alternative COALA Clusterings
Daiki Yamazaki, Mineichi Kudo,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) COALA, which is based on a hierarchical clustering, is a methodology to explore the 2nd clustering result which is different from 1st clustering result using cannot-link constraint. In this study, we extend the method so as to find more than one such an alternative clustering result. Furthermore, we consider a soft cannot-link constraint instead of COALA's hard cannot-link. In addition, we discuss the estimation of the cluster number by Gap statistics. The experiments show that our proposal method can find new alternative clustering results with good quality.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Multiple alternative clusterings / Hierarchical clustering / Cannot-link constraint / Estimation of the cluster number / Gap statistics
Paper # PRMU2017-86
Date of Issue 2017-10-05 (PRMU)

Conference Information
Committee PRMU
Conference Date 2017/10/12(2days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Shinichi Sato(NII)
Vice Chair Hironobu Fujiyoshi(Chubu Univ.) / Yoshihisa Ijiri(Omron)
Secretary Hironobu Fujiyoshi(AIST) / Yoshihisa Ijiri(NAIST)
Assistant Masato Ishii(NEC) / Yusuke Sugano(Osaka Univ.)

Paper Information
Registration To Technical Committee on Pattern Recognition and Media Understanding
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Multiple Alternative COALA Clusterings
Sub Title (in English)
Keyword(1) Multiple alternative clusterings
Keyword(2) Hierarchical clustering
Keyword(3) Cannot-link constraint
Keyword(4) Estimation of the cluster number
Keyword(5) Gap statistics
1st Author's Name Daiki Yamazaki
1st Author's Affiliation Hokkaido University(Hokkaido Univ.)
2nd Author's Name Mineichi Kudo
2nd Author's Affiliation Hokkaido University(Hokkaido Univ.)
Date 2017-10-13
Paper # PRMU2017-86
Volume (vol) vol.117
Number (no) PRMU-238
Page pp.pp.133-138(PRMU),
#Pages 6
Date of Issue 2017-10-05 (PRMU)