Publication at Faculty of Mathematics and Physics |
2008
Abstract
Annotation in the original language is: In this paper we deal with the problem of learning user preferences from user?s scoring of a small sample of objects with labels from a very small linearly ordered set. The main task of this process is to use these preferences for a top-k query, which delivers the user with an ordered list of k highest ranked objects.
We deal with a problem of many ties in the highest score. Two algorithms for learning objective and utility functions are presented.
We experiment and compare them to some classical data-mining methods. We use several measures (RMSE and rank correlations ...) to evaluate efficiency of these methods.