Browse wiki

Jump to: navigation, search
Learning to rank with (a lot of) word features
Abstract In this article we present Supervised SemaIn this article we present Supervised Semantic Indexing which defines a class of nonlinear (quadratic) models that are discriminatively trained to directly map from the word content in a query-document or document-document pair to a ranking score. Like Latent Semantic Indexing {(LSI),} our models take account of correlations between words (synonymy, polysemy). However, unlike {LSI} our models are trained from a supervised signal directly on the ranking task of interest, which we argue is the reason for our superior results. As the query and target texts are modeled separately, our approach is easily generalized to different retrieval tasks, such as cross-language retrieval or online advertising placement. Dealing with models on all pairs of words features is computationally challenging. We propose several improvements to our basic model for addressing this issue, including low rank (but diagonal preserving) representations, correlated feature hashing and sparsification. We provide an empirical study of all these methods on retrieval tasks based on Wikipedia documents as well as an Internet advertisement task. We obtain state-of-the-art performance while providing realistically scalable methods. providing realistically scalable methods.
Abstractsub In this article we present Supervised SemaIn this article we present Supervised Semantic Indexing which defines a class of nonlinear (quadratic) models that are discriminatively trained to directly map from the word content in a query-document or document-document pair to a ranking score. Like Latent Semantic Indexing {(LSI),} our models take account of correlations between words (synonymy, polysemy). However, unlike {LSI} our models are trained from a supervised signal directly on the ranking task of interest, which we argue is the reason for our superior results. As the query and target texts are modeled separately, our approach is easily generalized to different retrieval tasks, such as cross-language retrieval or online advertising placement. Dealing with models on all pairs of words features is computationally challenging. We propose several improvements to our basic model for addressing this issue, including low rank (but diagonal preserving) representations, correlated feature hashing and sparsification. We provide an empirical study of all these methods on retrieval tasks based on Wikipedia documents as well as an Internet advertisement task. We obtain state-of-the-art performance while providing realistically scalable methods. providing realistically scalable methods.
Bibtextype article  +
Has author Bing Bai + , Jason Weston + , David Grangier + , Ronan Collobert + , Kunihiko Sadamasa + , Yanjun Qi + , Olivier Chapelle + , Kilian Weinberger +
Has remote mirror http://0-portal.acm.org.mercury.concordia.ca/citation.cfm?id=1825381.1825399&coll=DL&dl=GUIDE&CFID=112025803&CFTOKEN=32336862&preflayout=flat  +
Number of citations by publication 0  +
Number of references by publication 0  +
Peer-reviewed Yes  +
Published in Information retrieval +
Title Learning to rank with (a lot of) word features +
Type journal article  +
Volume 13  +
Year 2010 +
Creation dateThis property is a special property in this wiki. 20 September 2014 12:49:31  +
Categories Publications without keywords parameter  + , Publications without language parameter  + , Publications without license parameter  + , Publications without DOI parameter  + , Publications without archive mirror parameter  + , Publications without paywall mirror parameter  + , Journal articles  + , Publications without references parameter  + , Publications  +
Modification dateThis property is a special property in this wiki. 20 September 2014 12:49:31  +
DateThis property is a special property in this wiki. 2010  +
hide properties that link here 
Learning to rank with (a lot of) word features + Title
 

 

Enter the name of the page to start browsing from.