Experiments in Bayesian Recommendation
Experiments in Bayesian Recommendation
The performance of collaborative filtering recommender systems can suffer when data is sparse, for example in distributed situations. In addition popular algorithms such as memory-based collaborative filtering are rather ad-hoc, making principled improvements difficult. In this paper we focus on a simple recommender based on naïve Bayesian techniques, and explore two different methods of modelling probabilities. We find that a Gaussian model for rating behaviour works well, and with the addition of a Gaussian-Gamma prior it maintains good performance even when data is sparse.
Recommender systems, Collaborative filtering, Bayesian methods, Naïve Bayes
978-3-642-18028-6
Barnard, Thomas
6e434c6b-7df9-4c2d-a9c3-3a95a8df9044
Prügel-Bennett, Adam
b107a151-1751-4d8b-b8db-2c395ac4e14e
26 January 2011
Barnard, Thomas
6e434c6b-7df9-4c2d-a9c3-3a95a8df9044
Prügel-Bennett, Adam
b107a151-1751-4d8b-b8db-2c395ac4e14e
Barnard, Thomas and Prügel-Bennett, Adam
(2011)
Experiments in Bayesian Recommendation.
In Proceedings of the 7th Atlantic Web Intelligence Conference, AWIC 2011, Fribourg, Switzerland, January 26-28, 2011.
Springer..
Record type:
Conference or Workshop Item
(Paper)
Abstract
The performance of collaborative filtering recommender systems can suffer when data is sparse, for example in distributed situations. In addition popular algorithms such as memory-based collaborative filtering are rather ad-hoc, making principled improvements difficult. In this paper we focus on a simple recommender based on naïve Bayesian techniques, and explore two different methods of modelling probabilities. We find that a Gaussian model for rating behaviour works well, and with the addition of a Gaussian-Gamma prior it maintains good performance even when data is sparse.
Text
paper3.pdf
- Accepted Manuscript
More information
Published date: 26 January 2011
Keywords:
Recommender systems, Collaborative filtering, Bayesian methods, Naïve Bayes
Organisations:
Southampton Wireless Group
Identifiers
Local EPrints ID: 271747
URI: http://eprints.soton.ac.uk/id/eprint/271747
ISBN: 978-3-642-18028-6
PURE UUID: db11833b-7c39-4ec4-b2c5-3cf728297de8
Catalogue record
Date deposited: 02 Dec 2010 23:46
Last modified: 14 Mar 2024 09:38
Export record
Contributors
Author:
Thomas Barnard
Author:
Adam Prügel-Bennett
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics