Hits:

Indexed by:期刊论文

Date of Publication:2014-12-01

Journal:Journal of Computational Information Systems

Included Journals:EI、Scopus

Volume:10

Issue:23

Page Number:10195-10202

ISSN No.:15539105

Abstract:Learning ranking functions from preference data in particular had recently attracted much interest. The ranking algorithms were often evaluated using information retrieval measures; the main difficulty in direct optimization of these measures was that they depended on the ranks of documents. The roles of preference were investigated between the relevant documents and irrelevant documents in the learning process. To remedy this, a new input sample named one-group sample was constructed by a relevant document and a group of irrelevant documents according to a given query. The new sample could distinguish the relevance of documents effectively; by the new samples two new loss function was also developed to improve the performance of learned ranking functions. Copyright ? 2014 Binary Information Press.