Bidding and allocation in combinatorial auctions
Proceedings of the 2nd ACM conference on Electronic commerce
Hybrid Recommender Systems: Survey and Experiments
User Modeling and User-Adapted Interaction
Solving concisely expressed combinatorial auction problems
Eighteenth national conference on Artificial intelligence
TiVo: making show recommendations using a distributed collaborative filtering architecture
Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining
Improving recommendation lists through topic diversification
WWW '05 Proceedings of the 14th international conference on World Wide Web
Combinatorial Auctions
Journal of Artificial Intelligence Research
Preference-based configuration of web page content
IJCAI'01 Proceedings of the 17th international joint conference on Artificial intelligence - Volume 2
Learning user preferences for sets of objects
ICML '06 Proceedings of the 23rd international conference on Machine learning
Heuristic search and information visualization methods for school redistricting
IAAI'06 Proceedings of the 18th conference on Innovative applications of artificial intelligence - Volume 2
AAAI'06 proceedings of the 21st national conference on Artificial intelligence - Volume 2
AAAI'07 Proceedings of the 22nd national conference on Artificial intelligence - Volume 2
Generic preferences over subsets of structured objects
Journal of Artificial Intelligence Research
Learning optimal subsets with implicit user preferences
IJCAI'09 Proceedings of the 21st international jont conference on Artifical intelligence
Representing and reasoning with qualitative preferences for compositional systems
Journal of Artificial Intelligence Research
Generating diverse plans to handle unknown and partially known user preferences
Artificial Intelligence
Hi-index | 0.00 |
In many application domains, it is useful to be able to represent and reason about a user's preferences over sets of objects. We present a representation language, DD-PREF (for Diversity and Depth PREFerences), for specifying the desired diversity and depth of sets of objects where each object is represented as a vector of feature values. A strong diversity preference for a particular feature indicates that the user would like the set to include objects whose values are evenly dispersed across the range of possible values for that feature. A strong depth preference for a feature indicates that the user is interested in specific target values or ranges. Diversity and depth are complementary, but are not necessarily opposites. We define an objective function that, when maximized, identifies the subset of objects that best satisfies a statement of preferences in DD-PREF. Exhaustively searching the space of all possible subsets is intractable for large problem spaces; therefore, we also present an efficient greedy algorithm for generating preferred object subsets. We demonstrate the expressive power of DD-PREF and the performance of our greedy algorithm by encoding and applying qualitatively different preferences for multiple tasks on a blocks world data set. Finally, we provide experimental results for a collection of Mars rover images, demonstrating that we can successfully capture individual preferences of different users, and use them to retrieve high-quality image subsets.