EFFECTS OF BINARY SIMILARITY MEASURES ON TOP-N RECOMMENDATIONS
Abstract
Shopping over the Internet through several e-commerce sites is receiving increasing attention. Customers want to purchase those products that they might like without wasting time and/or money. To help their customers, many online companies provide top-N recommendations by means of recommender systems. Similarity measures used to find out the most similar entities might affect the
overall performance of top-N predictions. Although there are various binary ratings-based similarity metrics, their effects on accuracy and online efficiency of top-N recommendations have not been deeply studied.
In this study, we investigate seven well-known binary ratings-based similarity metrics in terms of both preciseness and efficiency while providing top-N recommendations. To compare them with respect to accuracy and competence, we perform several experiments based on two well-known real data sets. We modify top-N recommendation algorithm in such a way so that the most similar users' data
are involved in recommendation process. We also study how varying controlling parameters affect overall performance with different similarity metrics. We analyze our empirical results and provide some suggestions.
overall performance of top-N predictions. Although there are various binary ratings-based similarity metrics, their effects on accuracy and online efficiency of top-N recommendations have not been deeply studied.
In this study, we investigate seven well-known binary ratings-based similarity metrics in terms of both preciseness and efficiency while providing top-N recommendations. To compare them with respect to accuracy and competence, we perform several experiments based on two well-known real data sets. We modify top-N recommendation algorithm in such a way so that the most similar users' data
are involved in recommendation process. We also study how varying controlling parameters affect overall performance with different similarity metrics. We analyze our empirical results and provide some suggestions.
Keywords
Similarity metric, Top-N recommendation, Accuracy, Online efficiency
References
- Billus, D. and Pazzani, M.J. (1998). Learning Collaborative Information Filters. Proceed- ings of the 15th International Conference on Machine Learning, Adison, WI, USA, 46- 54.
- Brožovský, L. (2006). Recommender System for A Dating Service. Master’s thesis, Prague, Czech Republic: Charles University in Prague.
- Cha, S.-H., Yoon, S., and Tappert, C.C. (2005). On Binary Similarity Measures for Hand- written Character Recognition. Proceedings of the 2005 8th International Conference on Document Analysis and Recognition, Seoul, Korea, vol. 1, 4-8.
- Choi, S.-S., Cha, S.-H., and Tappert, C.C. (2010). A Survey of Binary Similarity and Distance Measures. Journal of Systemics, Cybernetics and Informatics 8 (1), 43-48.
- GroupLens Research, Data Sets, 2006. http://www.grouplens.org/node/12. (May 15, 2012).
- Kaleli, C. and Polat, H. (2009). Similar or Dis- similar Users? Or both? Proceedings of the 2009 2nd International Symposium on Elec- tronic Commerce and Security, Nanchang City, China, vol. 2, 184-189.
- Miranda, C. and Jorge, A.M. (2009). Item-based and user-based incremental collaborative fil- tering for Web recommendations. Lecture Notes in Computer Science 5816, 673-684.
- Miyahara, K. and Pazzani, M.J. (2000). Collabo- rative filtering with the simple Bayesian clas- sifier. Lecture Notes in Computer Science 1886, 679-689.
- Miyahara, K. and Pazzani, M.J. (2002). Im- provement collaborative filtering with the simple Bayesian classifier. Transactions of Information Processing Society of Japan 43 (11), 3429-3437.
- Papagelis, M., Rousidis, I., Plexousakis, D., and Theoharopoulos, E. (2005). Incremental Col- laborative Filtering for Highly-Scalable Rec- ommendation Algorithms. Lecture Notes in Computer Science 3488, 553-561.