Social experiment on advisory recommender system for energy-saving

Authors:
Hiroki Shigeyoshi;Ken'iti Tamano;Ryosuke Saga;Hiroshi Tsuji;Shuki Inoue;Tsuyoshi Ueno
Affiliations:
Osaka Prefecture University, Sakai-shi, Osaka, Japan;Osaka Prefecture University, Sakai-shi, Osaka, Japan;Osaka Prefecture University, Sakai-shi, Osaka, Japan;Osaka Prefecture University, Sakai-shi, Osaka, Japan;Energy Use R&D Center, The Kansai Electric Power Co., Inc., Amagasaki-shi, Hyogo, Japan;Central Research Institute of Electric Power Industry, Komae-shi, Tokyo, Japan
Venue:
HCI International'13 Proceedings of the 15th international conference on Human Interface and the Management of Information: information and interaction design - Volume Part I
Year:
2013

Citing 8
Cited 0

Recommender systems

Communications of the ACM
Item-based collaborative filtering recommendation algorithms

Proceedings of the 10th international conference on World Wide Web
E-Commerce Recommendation Applications

Data Mining and Knowledge Discovery
Evaluating collaborative filtering recommender systems

ACM Transactions on Information Systems (TOIS)
Improving recommendation lists through topic diversification

WWW '05 Proceedings of the 14th international conference on World Wide Web
Recommender Systems Handbook

Recommender Systems Handbook
Knowledge and transaction based domestic energy saving support system

KES'11 Proceedings of the 15th international conference on Knowledge-based and intelligent information and engineering systems - Volume Part IV
Empirical analysis of predictive algorithms for collaborative filtering

UAI'98 Proceedings of the Fourteenth conference on Uncertainty in artificial intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper describes a social experiment on an advisory recommender system for home energy-saving, called KNOTES. Based on the user's value sense and the effectiveness of the advice, KNOTES aims to recommend highly effective advices over the user's own preferences. In addition, KNOTES uses an advice reference history to avoid the repetition of redundant advice. For the social experiment, forty-seven subjects used KNOTES for about two months. Introducing four metrics for comparing KNOTES with a random recommender, this paper verifies that KNOTES could recommend the advices which are desirable from the view of energy-saving and could avoid the repetition of redundant advices. The remaining issue has been prediction of the users' preferences according to their value sense.