GroupLens: applying collaborative filtering to Usenet news
Communications of the ACM
Recommendation as classification: using social and content-based information in recommendation
AAAI '98/IAAI '98 Proceedings of the fifteenth national/tenth conference on Artificial intelligence/Innovative applications of artificial intelligence
K-medians, facility location, and the Chernoff-Wald bound
SODA '00 Proceedings of the eleventh annual ACM-SIAM symposium on Discrete algorithms
Stated choice methods: analysis and application
Stated choice methods: analysis and application
The FindMe Approach to Assisted Browsing
IEEE Expert: Intelligent Systems and Their Applications
UCP-Networks: A Directed Graphical Representation of Conditional Utilities
UAI '01 Proceedings of the 17th Conference in Uncertainty in Artificial Intelligence
UAI '01 Proceedings of the 17th Conference in Uncertainty in Artificial Intelligence
A POMDP formulation of preference elicitation problems
Eighteenth national conference on Artificial intelligence
Journal of Artificial Intelligence Research
Trust building with explanation interfaces
Proceedings of the 11th international conference on Intelligent user interfaces
Increasing user decision accuracy using suggestions
Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
Learning user preferences for sets of objects
ICML '06 Proceedings of the 23rd international conference on Machine learning
Refining preference-based search results through Bayesian filtering
Proceedings of the 12th international conference on Intelligent user interfaces
International Journal of Electronic Commerce
Trust-inspiring explanation interfaces for recommender systems
Knowledge-Based Systems
Conversational recommenders with adaptive suggestions
Proceedings of the 2007 ACM conference on Recommender systems
marService: multiattribute utility recommendation for e-markets
International Journal of Computer Applications in Technology
Preference-based search using example-critiquing with suggestions
Journal of Artificial Intelligence Research
Generic preferences over subsets of structured objects
Journal of Artificial Intelligence Research
Regret-based optimal recommendation sets in conversational recommender systems
Proceedings of the third ACM conference on Recommender systems
Learning optimal subsets with implicit user preferences
IJCAI'09 Proceedings of the 21st international jont conference on Artifical intelligence
Experiments on the preference-based organization interface in recommender systems
ACM Transactions on Computer-Human Interaction (TOCHI)
A Survey of Accuracy Evaluation Metrics of Recommendation Tasks
The Journal of Machine Learning Research
Hi-index | 0.00 |
We propose an approach to recommendation systems that optimizes over possible sets of recommended alternatives in a decision-theoretic manner. Our approach selects the alternative set that maximizes the expected valuation of the user's choice from the recommended set. The set-based optimization explicitly recognizes the opportunity for passing residual uncertainty about preferences back to the user to resolve. Implicitly, the approach chooses a set with a diversity of alternatives that optimally covers the uncertainty over possible user preferences. The approach can be used with several preference representations, including utility theory, qualitative preferences models, and informal scoring. We develop a specific formulation for multi-attribute utility theory, which we call maximization of expected max (MEM). We go on to show that this optimization is NP-complete (when user preferences are described by discrete distributions) and suggest two efficient methods for approximating it. These approximations have complexity of the same order as the traditional k-max operator and, for both synthetic and real-world data, perform better than the approach of recommending the k-individually best alternatives (which is not a surprise) and very close to the optimum set (which is less expected).