Ordinal Boltzmann Machines for Collaborative Filtering

Tran The Truyen, Dinh Phung, Svetha Venkatesh
Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence, PMLR R7:556-564, 2009.

Abstract

Collaborative filtering is an effective recommendation technique wherein the preference of an individual can potentially be predicted based on preferences of other members. Early algorithms often relied on the strong locality in the preference data, that is, it is enough to predict preference of a user on a particular item based on a small subset of other users with similar tastes or of other items with similar properties. More recently, dimensionality reduction techniques have proved to be equally competitive, and these are based on the co-occurrence patterns rather than locality. This paper explores and extends a probabilistic model known as Boltzmann Machine for collaborative filtering tasks. It seamlessly integrates both the similarity and co-occurrence in a principled manner. In particular, we study parameterisation options to deal with the ordinal nature of the preferences, and propose a joint modelling of both the user-based and item-based processes. Experiments on moderate and large-scale movie recommendation show that our framework rivals existing well-known methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR7-truyen09a, title = {Ordinal {B}oltzmann Machines for Collaborative Filtering}, author = {Truyen, Tran The and Phung, Dinh and Venkatesh, Svetha}, booktitle = {Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence}, pages = {556--564}, year = {2009}, editor = {Bilmes, Jeff and Ng, Andrew Y.}, volume = {R7}, series = {Proceedings of Machine Learning Research}, month = {18--21 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r7/main/assets/truyen09a/truyen09a.pdf}, url = {https://proceedings.mlr.press/r7/truyen09a.html}, abstract = {Collaborative filtering is an effective recommendation technique wherein the preference of an individual can potentially be predicted based on preferences of other members. Early algorithms often relied on the strong locality in the preference data, that is, it is enough to predict preference of a user on a particular item based on a small subset of other users with similar tastes or of other items with similar properties. More recently, dimensionality reduction techniques have proved to be equally competitive, and these are based on the co-occurrence patterns rather than locality. This paper explores and extends a probabilistic model known as Boltzmann Machine for collaborative filtering tasks. It seamlessly integrates both the similarity and co-occurrence in a principled manner. In particular, we study parameterisation options to deal with the ordinal nature of the preferences, and propose a joint modelling of both the user-based and item-based processes. Experiments on moderate and large-scale movie recommendation show that our framework rivals existing well-known methods.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Ordinal Boltzmann Machines for Collaborative Filtering %A Tran The Truyen %A Dinh Phung %A Svetha Venkatesh %B Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2009 %E Jeff Bilmes %E Andrew Y. Ng %F pmlr-vR7-truyen09a %I PMLR %P 556--564 %U https://proceedings.mlr.press/r7/truyen09a.html %V R7 %X Collaborative filtering is an effective recommendation technique wherein the preference of an individual can potentially be predicted based on preferences of other members. Early algorithms often relied on the strong locality in the preference data, that is, it is enough to predict preference of a user on a particular item based on a small subset of other users with similar tastes or of other items with similar properties. More recently, dimensionality reduction techniques have proved to be equally competitive, and these are based on the co-occurrence patterns rather than locality. This paper explores and extends a probabilistic model known as Boltzmann Machine for collaborative filtering tasks. It seamlessly integrates both the similarity and co-occurrence in a principled manner. In particular, we study parameterisation options to deal with the ordinal nature of the preferences, and propose a joint modelling of both the user-based and item-based processes. Experiments on moderate and large-scale movie recommendation show that our framework rivals existing well-known methods. %Z Reissued by PMLR on 04 October 2026.
APA
Truyen, T.T., Phung, D. & Venkatesh, S.. (2009). Ordinal Boltzmann Machines for Collaborative Filtering. Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R7:556-564 Available from https://proceedings.mlr.press/r7/truyen09a.html. Reissued by PMLR on 04 October 2026.

Related Material