Conjugate-prior-regularized multinomial pLSA for collaborative filtering

Marcus Klasson, Stefan Ingi Adalbjörnsson, Johan Swärd, Søren Vang Andersen

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceedingpeer-review

Abstract

We consider the over-fitting problem for multinomial probabilistic Latent Semantic Analysis (pLSA) in collaborative filtering, using a regularization approach. For big data applications, the computational complexity is at a premium and we, therefore, consider a maximum a posteriori approach based on conjugate priors that ensure that complexity of each step remains the same as compared to the un-regularized method. In the numerical section, we show that the proposed regularization method and training scheme yields an improvement on commonly used data sets, as compared to previously proposed heuristics.

Original languageEnglish
Title of host publication25th European Signal Processing Conference, EUSIPCO 2017
PublisherIEEE - Institute of Electrical and Electronics Engineers Inc.
Pages2501-2505
Number of pages5
Volume2017-January
ISBN (Electronic)9780992862671
DOIs
Publication statusPublished - 2017 Oct 23
Event25th European Signal Processing Conference, EUSIPCO 2017 - Kos island, Kos, Greece
Duration: 2017 Aug 282017 Sept 2

Conference

Conference25th European Signal Processing Conference, EUSIPCO 2017
Country/TerritoryGreece
CityKos
Period2017/08/282017/09/02

Subject classification (UKÄ)

  • Signal Processing

Free keywords

  • Collaborative filtering
  • Conjugate prior regularization
  • Probabilistic latent semantic analysis
  • Recommender systems

Fingerprint

Dive into the research topics of 'Conjugate-prior-regularized multinomial pLSA for collaborative filtering'. Together they form a unique fingerprint.

Cite this