Learning to rank: from pairwise approach to listwise approach

DMCA

Learning to rank: from pairwise approach to listwise approach (2007)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Zhe Cao , Tao Qin , Tie-yan Liu , Hang Li

Venue:	In Proc. ICML’07
Citations:	248 - 30 self

BibTeX

@INPROCEEDINGS{Cao07learningto,
    author = {Zhe Cao and Tao Qin and Tie-yan Liu and Hang Li},
    title = {Learning to rank: from pairwise approach to listwise approach},
    booktitle = {In Proc. ICML’07},
    year = {2007},
    pages = {129--136}
}

OpenURL

Abstract

The paper is concerned with learning to rank, which is to construct a model or a function for ranking objects. Learning to rank is useful for document retrieval, collaborative filtering, and many other applications. Several methods for learning to rank have been proposed, which take object pairs as ‘instances ’ in learning. We refer to them as the pairwise approach in this paper. Although the pairwise approach offers advantages, it ignores the fact that ranking is a prediction task on list of objects. The paper postulates that learning to rank should adopt the listwise approach in which lists of objects are used as ‘instances ’ in learning. The paper proposes a new probabilistic method for the approach. Specifically it introduces two probability models, respectively referred to as permutation probability and top one probability, to define a listwise loss function for learning. Neural Network and Gradient Descent are then employed as model and algorithm in the learning method. Experimental results on information retrieval show that the proposed listwise approach performs better than the pairwise approach. Microsoft technique report. A short version of this work is published

Keyphrases

pairwise approach listwise approach short version collaborative filtering many application document retrieval new probabilistic method microsoft technique report object pair gradient descent listwise loss function permutation probability neural network several method experimental result information retrieval show pairwise approach offer advantage prediction task learning method probability model

DMCA