AUTHORS: Giang Nguyen Trung, Heejune Ahn

Download as PDF

ABSTRACT: The demand for recommendation systems gets higher and more popular these days. There have been several studies on recommendation systems, but the most recent ones give focus on the product as a whole and do not give much attention to the user's preferences such as price, type, and color. This paper proposes prefiltering methods and compares the benefits and performance between pre- and post-filtering methods. The prefiltering method ignores ratings of items that are not relevant to the user’s preferences, then reduces the size of target data set to process, saving processing time. The experimental result with MovieLens dataset shows that pre-filtering can provide the recommendation with 8.5 times less computations than post-filtering by restricting item set, and shows 2% improvement in F measurement. Moreover, rating estimation performance can vary from 1% improvement in the ML-1M dataset to 1% decrease in the ML-100K dataset in the RMSE

KEYWORDS: Information Retrieval; Conditional Recommendation; Matrix Factorization; Hierarchical System; Recommendation System; Pre-filtering

REFERENCES:

[1] J. Bennett, S. Lanning, The netflix prize, in: Proceedings of KDD cup and workshop, 2007, p. 35.

[2] G. Adomavicius, A. Tuzhilin, Context-aware recommender systems, Recommender systems handbook Springer US (2015) 191-226.

[3] P. Kouki, S. Fakhraei, J. Foulds, M. Eirinaki, L. Getoor, HyPER: A Flexible and Extensible Probabilistic Framework for Hybrid Recommender Systems, in: Proceedings of the 9th ACM Conference on Recommender Systems, 2015, pp. 99-106.

[4] J. McAuley, J. Leskovec, Hidden factors and hidden topics: Understanding rating dimensions with review text, in: Proceedings of the 7th ACM conference on Recommender systems, 2013, pp. 165-172

[5] J. Nguyen, M. Zhu, Content-boosted matrix factorization techniques for recommender systems, Statistical Analysis and Data Mining 6 (4) (2013) 286-301.

[6] G. Adomavicius, A. Tuzhilin, Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions, IEEE Transactions on Knowledge and Data Engineering 17 (6) (2005) 734–749.

[7] F. Ricci, L. Rokach, B. Shapira, P. B. Kantor , Recommender Systems Handbook, Springer, 2010.

[8] Y. Koren, R. Bell, C. Volinsky, Matrix factorization techniques for recommender systems, Computer 42 (8) (2009) 30-37.

[9] G. Adomavicius, R. Sankaranarayanan, S. Sen, A. Tuzhilin, Incorporating contextual information in recommender systems using a multidimensional approach, ACM Transactions on Information Systems 23 (1) (2005) 103-145.

[10] G. Guo, J. Zhang, Z. Sun, N. Yorke-Smith, LibRec: A Java Library for Recommender Systems, in: Proceedings of the 23rd Conference on User Modelling, Adaptation and Personalization,