Presentation 2012-11-07
sNMF: Stacked Non-negative Matrix Factorization : Its Application For Social Media Analysis
Koh TAKEUCHI, Katsuhiko ISHIGURO, Akisato KIMURA, Hiroshi SAWADA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper proposes a new technique named Stacked Non-negative Matrix Factorization (sNMF), which factorizes multiple non-negative matrices that are mutually related. The model is formulated as a generalization of the standard NMF, and also can be seen as a special case of NMF with undefined sub matrices. In this paper, we apply sNMF to an emerging social media analysis. More specifically, we employ sNMF to detect topics and find credible user clusters behind social media dataset. Experimental results show that sNMF performs much better than the standard NMF quantitatively, and successfully reveal reasonable topics and user clusters from social media dataset.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) NMF / Topic Model / Social Media Analysis
Paper # IBISML2012-43
Date of Issue

Conference Information
Committee IBISML
Conference Date 2012/10/31(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 Information-Based Induction Sciences and Machine Learning (IBISML)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) sNMF: Stacked Non-negative Matrix Factorization : Its Application For Social Media Analysis
Sub Title (in English)
Keyword(1) NMF
Keyword(2) Topic Model
Keyword(3) Social Media Analysis
1st Author's Name Koh TAKEUCHI
1st Author's Affiliation NTT Communication Science Laboratories, NTT Corporation()
2nd Author's Name Katsuhiko ISHIGURO
2nd Author's Affiliation NTT Communication Science Laboratories, NTT Corporation
3rd Author's Name Akisato KIMURA
3rd Author's Affiliation NTT Communication Science Laboratories, NTT Corporation
4th Author's Name Hiroshi SAWADA
4th Author's Affiliation NTT Communication Science Laboratories, NTT Corporation
Date 2012-11-07
Paper # IBISML2012-43
Volume (vol) vol.112
Number (no) 279
Page pp.pp.-
#Pages 8
Date of Issue