A HYBRID GA-BASED COLLABORATIVE FILTERING MODEL FOR ONLINE RECOMMENDERS

Yvonne Ho, Simon Fong, Zhuang Yan

2007

Abstract

Online recommenders have been a prevalent tool in e-Business that assists users to find items of their interest. Different algorithms that predict scores based on the heuristic of similarity of other peoples’ taste to that of the user have been experimented by many researchers. However most of the current recommendation systems take all users with common items as neighbors in their measurement, which may induce noise and hence inaccurate prediction. In this paper, we attempt to solve this problem by proposing a hybrid model that combines content-based filtering and collaborative filtering for online recommenders. The model exploits merits from these two techniques by selectively encoding both the user profiles and the product information into the same chromosomes in a Genetic Algorithm. Our experiments demonstrated that this new approach gives relatively high accuracy rate in predicting user preferences.

References

  1. Breese, J., Heckerman, D., and Kadie, C., 1998.Empirical Analysis of Predictive Algorithms for Collaborative Filtering. Proceedings of the 14th Conference on Uncertainty in Artificial Intelligence.
  2. Ujjin, S., Bentley, P., 2002. Evolving Good Recommendations. Genetic and Evolutionary Computation Conference (GECCO).
Download


Paper Citation


in Harvard Style

Ho Y., Fong S. and Yan Z. (2007). A HYBRID GA-BASED COLLABORATIVE FILTERING MODEL FOR ONLINE RECOMMENDERS . In Proceedings of the Second International Conference on e-Business - Volume 1: ICE-B, (ICETE 2007) ISBN 978-989-8111-11-1, pages 200-203. DOI: 10.5220/0002113602000203


in Bibtex Style

@conference{ice-b07,
author={Yvonne Ho and Simon Fong and Zhuang Yan},
title={A HYBRID GA-BASED COLLABORATIVE FILTERING MODEL FOR ONLINE RECOMMENDERS},
booktitle={Proceedings of the Second International Conference on e-Business - Volume 1: ICE-B, (ICETE 2007)},
year={2007},
pages={200-203},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002113602000203},
isbn={978-989-8111-11-1},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Second International Conference on e-Business - Volume 1: ICE-B, (ICETE 2007)
TI - A HYBRID GA-BASED COLLABORATIVE FILTERING MODEL FOR ONLINE RECOMMENDERS
SN - 978-989-8111-11-1
AU - Ho Y.
AU - Fong S.
AU - Yan Z.
PY - 2007
SP - 200
EP - 203
DO - 10.5220/0002113602000203