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

J Wang; AP de Vries; MJT Reinders

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

J Wang, AP de Vries, MJT Reinders

Multimedia Computing

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

67 Citations (Scopus)

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.

Original language	Undefined/Unknown
Title of host publication	Advances in information retrieval
Editors	M Lalmas, A MacFarlane, S Rüger, A Tombros, T Tsikrika, A Yavlinsky
Place of Publication	Heidelberg
Publisher	Springer
Pages	37-48
Number of pages	12
Publication status	Published - 2006
Event	28th European conference on IR research, ECIR 2006, London, UK - Heidelberg Duration: 10 Apr 2006 → 12 Apr 2006

Publication series

Name
Publisher	Springer

Name	Lecture Notes in Computer Science
Volume	3936
ISSN (Print)	0302-9743

Conference

Conference	28th European conference on IR research, ECIR 2006, London, UK
Period	10/04/06 → 12/04/06

Keywords

Wiskunde en Informatica
Techniek
technische Wiskunde en Informatica
conference contrib. refereed
CWTS 0.75 <= JFIS < 2.00

Cite this

@inproceedings{77877bb361ba4d6c835b7ecdf172f071,

title = "A user-item relevance model for log-based collaborative filtering",

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.",

keywords = "Wiskunde en Informatica, Techniek, technische Wiskunde en Informatica, conference contrib. refereed, CWTS 0.75 <= JFIS < 2.00",

author = "J Wang and {de Vries}, AP and MJT Reinders",

year = "2006",

language = "Undefined/Unknown",

publisher = "Springer",

pages = "37--48",

editor = "M Lalmas and A MacFarlane and S R{\"u}ger and A Tombros and T Tsikrika and A Yavlinsky",

booktitle = "Advances in information retrieval",

note = "28th European conference on IR research, ECIR 2006, London, UK ; Conference date: 10-04-2006 Through 12-04-2006",

}

Wang, J, de Vries, AP & Reinders, MJT 2006, A user-item relevance model for log-based collaborative filtering. in M Lalmas, A MacFarlane, S Rüger, A Tombros, T Tsikrika & A Yavlinsky (eds), Advances in information retrieval. Lecture Notes in Computer Science, vol. 3936, Springer, Heidelberg, pp. 37-48, 28th European conference on IR research, ECIR 2006, London, UK, 10/04/06.

A user-item relevance model for log-based collaborative filtering. / Wang, J; de Vries, AP; Reinders, MJT.
Advances in information retrieval. ed. / M Lalmas; A MacFarlane; S Rüger; A Tombros; T Tsikrika; A Yavlinsky. Heidelberg: Springer, 2006. p. 37-48 (Lecture Notes in Computer Science; Vol. 3936).

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - A user-item relevance model for log-based collaborative filtering

AU - Wang, J

AU - de Vries, AP

AU - Reinders, MJT

PY - 2006

Y1 - 2006

N2 - .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.

AB - .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.

KW - Wiskunde en Informatica

KW - Techniek

KW - technische Wiskunde en Informatica

KW - conference contrib. refereed

KW - CWTS 0.75 <= JFIS < 2.00

UR - http://www.springerlink.com/content/0773871142698p68/fulltext.pdf

M3 - Conference contribution

SP - 37

EP - 48

BT - Advances in information retrieval

A2 - Lalmas, M

A2 - MacFarlane, A

A2 - Rüger, S

A2 - Tombros, A

A2 - Tsikrika, T

A2 - Yavlinsky, A

PB - Springer

CY - Heidelberg

T2 - 28th European conference on IR research, ECIR 2006, London, UK

Y2 - 10 April 2006 through 12 April 2006

ER -

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

Abstract

Publication series

Conference

Keywords

Other files and links

Cite this