A user-item relevance model for log-based collaborative filtering

None, None; None, None; None, None

A user-item relevance model for log-based collaborative filtering

Conference Paper (2006)

Author(s)

J Wang (TU Delft - Electrical Engineering, Mathematics and Computer Science)

AP de Vries (TU Delft - Electrical Engineering, Mathematics and Computer Science)

MJT Reinders (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Research Group

Multimedia Computing

CWTS 0.75 <= JFIS < 2.00 Techniek Conference contrib. refereed Wiskunde en Informatica Technische Wiskunde en Informatica

To reference this document use

https://resolver.tudelft.nl/uuid:77877bb3-61ba-4d6c-835b-7ecdf172f071

More Info

expand_more

Publication Year

2006

Research Group

Multimedia Computing

Pages (from-to)

37-48

Publisher

Springer

Event

28th European conference on IR research, ECIR 2006, London, UK (2006-04-10 - 2006-04-12), Heidelberg

Downloads counter

140

Abstract

.Implicitacquisitionofuserpreferencesmakeslog-basedcollaborative¿lteringfavorableinpracticetoaccomplishrecommendations.Inthispaper,wefollowaformalapproachintextretrievaltore-formulatetheproblem.Basedontheclassicprobabilityrankingprinciple,weproposeaprobabilisticuser-itemrelevancemodel.Underthisformalmodel,weshowthatuser-basedanditem-basedapproachesareonlytwodi¿erentfactorizationswithdi¿erentindependenceassumptions.Moreover,weshowthatsmoothingisanimportantaspecttoestimatetheparametersofthemodelsduetodatasparsity.Byaddinglinearinterpolationsmoothing,theproposedmodelgivesaprobabilisticjusti¿cationofusingTF×IDF-likeitemrankingincollaborative¿ltering.Besidesgivingtheinsightunderstandingoftheproblemofcollaborative¿ltering,wealsoshowexperimentsinwhichtheproposedmethodprovidesabetterrecommendationperformanceonamusicplay-listdataset.

No files available

Metadata only record. There are no files for this record.