Presentation 2016-01-15
Multi-stage Collaborative Filtering for Composite Service Recommendation
Taketo Sasaki, Yohei Murakami, Toru Ishida,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) There have been attempts to apply collaborative filtering to service recommendation. Most all of them are for atomic service by collaborative filtering using explicit evaluation value by users. However, in real world, it is difficult to get explicit feedback from users, and there are composite services in addition to atomic services. In this paper, we propose multi-stage collaborative filtering using invocation count as evaluation value in order to prevent users from giving feedback and to take account of service composition in composite services. Focusing on nested structure in composite service, our method make evaluation value matrix for each set of services that might be invoked by composite service, and apply collaborative filtering repeatedly until executable services are recommended to user. We applied proposed method to service recommendation on the Language Grid, and verified the effectiveness of multi-stage collaborative filtering and measured recommendation accuracy.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Service Recommendation / Composite Services / Collaborative Filtering
Paper # SC2015-21
Date of Issue 2016-01-08 (SC)

Conference Information
Committee SC
Conference Date 2016/1/15(1days)
Place (in Japanese) (See Japanese page)
Place (in English) Kyoto University (Design Innovation Center)
Topics (in Japanese) (See Japanese page)
Topics (in English) Cloud, IoT, and Service Innovation
Chair Shigeru Hosono(NEC)
Vice Chair Incheon Paik(Univ. of Aizu)
Secretary Incheon Paik(NICT)
Assistant

Paper Information
Registration To Technical Committee on Service Computing
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Multi-stage Collaborative Filtering for Composite Service Recommendation
Sub Title (in English)
Keyword(1) Service Recommendation
Keyword(2) Composite Services
Keyword(3) Collaborative Filtering
1st Author's Name Taketo Sasaki
1st Author's Affiliation Kyoto University(Kyoto Univ.)
2nd Author's Name Yohei Murakami
2nd Author's Affiliation Kyoto University(Kyoto Univ.)
3rd Author's Name Toru Ishida
3rd Author's Affiliation Kyoto University(Kyoto Univ.)
Date 2016-01-15
Paper # SC2015-21
Volume (vol) vol.115
Number (no) SC-393
Page pp.pp.37-42(SC),
#Pages 6
Date of Issue 2016-01-08 (SC)