NASA Logo

NTRS

NTRS - NASA Technical Reports Server

Back to Results
learning user preferences for sets of objectsMost work on preference learning has focused on pairwise preferences or rankings over individual items. In this paper, we present a method for learning preferences over sets of items. Our learning method takes as input a collection of positive examples--that is, one or more sets that have been identified by a user as desirable. Kernel density estimation is used to estimate the value function for individual items, and the desired set diversity is estimated from the average set diversity observed in the collection. Since this is a new learning problem, we introduce a new evaluation methodology and evaluate the learning method on two data collections: synthetic blocks-world data and a new real-world music data collection that we have gathered.
Document ID
20060050176
Document Type
Preprint (Draft being sent to journal)
External Source(s)
Authors
desJardins, Marie
(Maryland Univ. Baltimore County Baltimore, MD, United States)
Eaton, Eric
(Maryland Univ. Baltimore County Baltimore, MD, United States)
Wagstaff, Kiri L.
(Jet Propulsion Lab., California Inst. of Tech. Pasadena, CA, United States)
Date Acquired
August 23, 2013
Publication Date
January 1, 2006
Subject Category
Mathematical and Computer Sciences (General)
Meeting Information
23rd International Conference on Machine Learning(Pittsburgh, PA)
Funding Number(s)
CONTRACT_GRANT: NSF 03-25329
Distribution Limits
Public
Copyright
Other
Keywords
preferences
subset selections