Preference Networks: Probabilistic Models for Recommendation Systems

TL;DR

Preference Networks provide a unified probabilistic framework that integrates content-based and collaborative filtering for recommendation systems, enabling flexible predictions and queries.

Contribution

This paper introduces Preference Networks, a novel probabilistic model combining content and collaborative filtering into a single framework for recommendation.

Findings

Effective in rating prediction and top-N recommendation

Handles large user-item networks with pseudo-likelihood learning

Demonstrates superior performance on movie rating data

Abstract

Recommender systems are important to help users select relevant and personalised information over massive amounts of data available. We propose an unified framework called Preference Network (PN) that jointly models various types of domain knowledge for the task of recommendation. The PN is a probabilistic model that systematically combines both content-based filtering and collaborative filtering into a single conditional Markov random field. Once estimated, it serves as a probabilistic database that supports various useful queries such as rating prediction and top-$N$ recommendation. To handle the challenging problem of learning large networks of users and items, we employ a simple but effective pseudo-likelihood with regularisation. Experiments on the movie rating data demonstrate the merits of the PN.