A non-IID framework for collaborative filtering with Restricted Boltzmann Machines

Kostadin Georgiev, Preslav Nakov

Research output: Chapter in Book/Report/Conference proceedingConference contribution

49 Citations (Scopus)

Abstract

We propose a framework for collaborative filtering based on Restricted Boltzmann Machines (RBM), which extends previous RBM-based approaches in several important directions. First, while previous RBM research has focused on modeling the correlation between item ratings, we model both user-user and item-item correlations in a unified hybrid non-IID framework. We further use real values in the visible layer as opposed to multinomial variables, thus taking advantage of the natural order between user-item ratings. Finally, we explore the potential of combining the original training data with data generated by the RBM-based model itself in a bootstrapping fashion. The evaluation on two MovieLens datasets (with 100K and 1M user-item ratings, respectively), shows that our RBM model rivals the best previously-proposed approaches.

Original languageEnglish
Title of host publication30th International Conference on Machine Learning, ICML 2013
PublisherInternational Machine Learning Society (IMLS)
Pages2185-2193
Number of pages9
EditionPART 3
Publication statusPublished - 1 Jan 2013
Event30th International Conference on Machine Learning, ICML 2013 - Atlanta, GA, United States
Duration: 16 Jun 201321 Jun 2013

Other

Other30th International Conference on Machine Learning, ICML 2013
CountryUnited States
CityAtlanta, GA
Period16/6/1321/6/13

Fingerprint

Collaborative filtering
rating
evaluation

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Sociology and Political Science

Cite this

Georgiev, K., & Nakov, P. (2013). A non-IID framework for collaborative filtering with Restricted Boltzmann Machines. In 30th International Conference on Machine Learning, ICML 2013 (PART 3 ed., pp. 2185-2193). International Machine Learning Society (IMLS).

A non-IID framework for collaborative filtering with Restricted Boltzmann Machines. / Georgiev, Kostadin; Nakov, Preslav.

30th International Conference on Machine Learning, ICML 2013. PART 3. ed. International Machine Learning Society (IMLS), 2013. p. 2185-2193.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Georgiev, K & Nakov, P 2013, A non-IID framework for collaborative filtering with Restricted Boltzmann Machines. in 30th International Conference on Machine Learning, ICML 2013. PART 3 edn, International Machine Learning Society (IMLS), pp. 2185-2193, 30th International Conference on Machine Learning, ICML 2013, Atlanta, GA, United States, 16/6/13.
Georgiev K, Nakov P. A non-IID framework for collaborative filtering with Restricted Boltzmann Machines. In 30th International Conference on Machine Learning, ICML 2013. PART 3 ed. International Machine Learning Society (IMLS). 2013. p. 2185-2193
Georgiev, Kostadin ; Nakov, Preslav. / A non-IID framework for collaborative filtering with Restricted Boltzmann Machines. 30th International Conference on Machine Learning, ICML 2013. PART 3. ed. International Machine Learning Society (IMLS), 2013. pp. 2185-2193
@inproceedings{af5074a5e13a4579a1bd436cab4e9f3a,
title = "A non-IID framework for collaborative filtering with Restricted Boltzmann Machines",
abstract = "We propose a framework for collaborative filtering based on Restricted Boltzmann Machines (RBM), which extends previous RBM-based approaches in several important directions. First, while previous RBM research has focused on modeling the correlation between item ratings, we model both user-user and item-item correlations in a unified hybrid non-IID framework. We further use real values in the visible layer as opposed to multinomial variables, thus taking advantage of the natural order between user-item ratings. Finally, we explore the potential of combining the original training data with data generated by the RBM-based model itself in a bootstrapping fashion. The evaluation on two MovieLens datasets (with 100K and 1M user-item ratings, respectively), shows that our RBM model rivals the best previously-proposed approaches.",
author = "Kostadin Georgiev and Preslav Nakov",
year = "2013",
month = "1",
day = "1",
language = "English",
pages = "2185--2193",
booktitle = "30th International Conference on Machine Learning, ICML 2013",
publisher = "International Machine Learning Society (IMLS)",
edition = "PART 3",

}

TY - GEN

T1 - A non-IID framework for collaborative filtering with Restricted Boltzmann Machines

AU - Georgiev, Kostadin

AU - Nakov, Preslav

PY - 2013/1/1

Y1 - 2013/1/1

N2 - We propose a framework for collaborative filtering based on Restricted Boltzmann Machines (RBM), which extends previous RBM-based approaches in several important directions. First, while previous RBM research has focused on modeling the correlation between item ratings, we model both user-user and item-item correlations in a unified hybrid non-IID framework. We further use real values in the visible layer as opposed to multinomial variables, thus taking advantage of the natural order between user-item ratings. Finally, we explore the potential of combining the original training data with data generated by the RBM-based model itself in a bootstrapping fashion. The evaluation on two MovieLens datasets (with 100K and 1M user-item ratings, respectively), shows that our RBM model rivals the best previously-proposed approaches.

AB - We propose a framework for collaborative filtering based on Restricted Boltzmann Machines (RBM), which extends previous RBM-based approaches in several important directions. First, while previous RBM research has focused on modeling the correlation between item ratings, we model both user-user and item-item correlations in a unified hybrid non-IID framework. We further use real values in the visible layer as opposed to multinomial variables, thus taking advantage of the natural order between user-item ratings. Finally, we explore the potential of combining the original training data with data generated by the RBM-based model itself in a bootstrapping fashion. The evaluation on two MovieLens datasets (with 100K and 1M user-item ratings, respectively), shows that our RBM model rivals the best previously-proposed approaches.

UR - http://www.scopus.com/inward/record.url?scp=84897544713&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897544713&partnerID=8YFLogxK

M3 - Conference contribution

SP - 2185

EP - 2193

BT - 30th International Conference on Machine Learning, ICML 2013

PB - International Machine Learning Society (IMLS)

ER -