Presentation 2011-09-21
Support Selection Algorithms based on Collaborative Filtering for Adaptive Task Support Systems
Hideki AOYAMA, Kazuaki KONDO, Yuichi NAKAMURA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper presents a novel idea of utilizing recommendation algorithms, for support selection in a task support system. In this framework, appropriate supports can be selected based on ratings by collaborative filtering, considering one action of support as one item of recommendation. And we discuss differences between task support systems and recommendation systems : the amount of rating error is critical ; the performance should be measured for the whole supports. Therefore, we propose novel collaborative filtering algorithms robust to rating error and an algorithm to select supports appropriate as the whole task supports. We conducted experiments on performance evaluation of proposed algorithms in a cooking support environment. The results showed that an algorithm composed of Item-based Collaborative Filtering and Average method gives good performance, and probing effect improves the performances of most of the algorithms.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) task support system / method of support selection / collaborative filtering / probing effect / human computer interaction
Paper # SIS2011-25
Date of Issue

Conference Information
Committee SIS
Conference Date 2011/9/14(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 Smart Info-Media Systems (SIS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Support Selection Algorithms based on Collaborative Filtering for Adaptive Task Support Systems
Sub Title (in English)
Keyword(1) task support system
Keyword(2) method of support selection
Keyword(3) collaborative filtering
Keyword(4) probing effect
Keyword(5) human computer interaction
1st Author's Name Hideki AOYAMA
1st Author's Affiliation Academic Center for Computing and Media Studies, Kyoto University()
2nd Author's Name Kazuaki KONDO
2nd Author's Affiliation Academic Center for Computing and Media Studies, Kyoto University
3rd Author's Name Yuichi NAKAMURA
3rd Author's Affiliation Academic Center for Computing and Media Studies, Kyoto University
Date 2011-09-21
Paper # SIS2011-25
Volume (vol) vol.111
Number (no) 210
Page pp.pp.-
#Pages 6
Date of Issue