Kernel mapping recommender system algorithms
Kernel mapping recommender system algorithms
Recommender systems apply machine learning techniques for filtering unseen information and can predict whether a user would like a given item. In this paper, we propose a new algorithm that we call the Kernel-Mapping Recommender (KMR), which uses a novel structure learning technique. This paper makes the following contributions: we show how (1) user-based and item-based versions of the KMR algorithm can be built; (2) user-based and item-based versions can be combined; (3) more information—features, genre, etc.—can be employed using kernels and how this affects the final results; and (4) to make reliable recommendations under sparse, cold-start, and long tail scenarios. By extensive experimental results on five different datasets, we show that the proposed algorithms outperform or give comparable results to other state-of-the-art algorithms.
recommender systems, structure learning, linear operation, maximum margin, kernel
81-104
Ghazanfar, Mustansar
9452876d-46e5-4c7d-9120-c486f448632d
Prugel-Bennett, Adam
b107a151-1751-4d8b-b8db-2c395ac4e14e
Szedmak, Sandor
c6a84aa3-2956-4acf-8293-a1b676f6d7d8
November 2012
Ghazanfar, Mustansar
9452876d-46e5-4c7d-9120-c486f448632d
Prugel-Bennett, Adam
b107a151-1751-4d8b-b8db-2c395ac4e14e
Szedmak, Sandor
c6a84aa3-2956-4acf-8293-a1b676f6d7d8
Ghazanfar, Mustansar, Prugel-Bennett, Adam and Szedmak, Sandor
(2012)
Kernel mapping recommender system algorithms.
Information Sciences, 208, .
(doi:10.1016/j.ins.2012.04.012).
Abstract
Recommender systems apply machine learning techniques for filtering unseen information and can predict whether a user would like a given item. In this paper, we propose a new algorithm that we call the Kernel-Mapping Recommender (KMR), which uses a novel structure learning technique. This paper makes the following contributions: we show how (1) user-based and item-based versions of the KMR algorithm can be built; (2) user-based and item-based versions can be combined; (3) more information—features, genre, etc.—can be employed using kernels and how this affects the final results; and (4) to make reliable recommendations under sparse, cold-start, and long tail scenarios. By extensive experimental results on five different datasets, we show that the proposed algorithms outperform or give comparable results to other state-of-the-art algorithms.
Text
MMMR_IS_temp.pdf
- Other
Text
MMMR_IS_Finalized.pdf
- Other
More information
e-pub ahead of print date: 5 May 2012
Published date: November 2012
Keywords:
recommender systems, structure learning, linear operation, maximum margin, kernel
Organisations:
Southampton Wireless Group
Identifiers
Local EPrints ID: 272686
URI: http://eprints.soton.ac.uk/id/eprint/272686
ISSN: 0020-0255
PURE UUID: 90f98920-47ad-4096-b834-8544043d8c78
Catalogue record
Date deposited: 21 Aug 2011 14:30
Last modified: 14 Mar 2024 10:07
Export record
Altmetrics
Contributors
Author:
Mustansar Ghazanfar
Author:
Adam Prugel-Bennett
Author:
Sandor Szedmak
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