Abstract
User-based collaborative filtering, a widely used nearest neighbour-based recommendation technique, predicts an item’s rating by aggregating its ratings from similar users. User similarity is traditionally calculated by cosine similarity or the Pearson correlation coefficient. However, both of these measures consider only the direction of rating vectors, and suffer from a range of drawbacks. To overcome these issues, we propose a novel Bayesian similarity measure based on the Dirichlet distribution, taking into consideration both the direction and length of rating vectors. We posit that not all the rating pairs should be equally counted in order to accurately model user correlation. Three different evidence factors are designed to compute the weights of rating pairs. Further, our principled method reduces correlation due to chance and potential system bias. Experimental results on six real-world datasets show that our method achieves superior accuracy in comparison with counterparts.
- G. Adomavicius and J. Zhang. 2012. Impact of data characteristics on recommender systems performance. ACM Transactions on Management Information Systems 3, 1, 3. Google Scholar
Digital Library
- H. J. Ahn. 2008. A new similarity measure for collaborative filtering to alleviate the new user cold-starting problem. Information Sciences 178, 1, 37--51. Google Scholar
Digital Library
- A. Anderson, D. Huttenlocher, J. Kleinberg, and J. Leskovec. 2012. Effects of user similarity in social media. In Proceedings of the 5th ACM International Conference on Web Search and Data Mining (WSDM’12). 703--712. Google Scholar
Digital Library
- J. Bobadilla, F. Ortega, and A. Hernando. 2012. A collaborative filtering similarity measure based on singularities. Information Processing and Management 48, 2, 204--217. Google Scholar
Digital Library
- J. S. Breese, D. Heckerman, C. Kadie, and others. 1998. Empirical analysis of predictive algorithms for collaborative filtering. In Proceedings of the 14th Conference on Uncertainty in Artificial Intelligence (UAI’98). 43--52. Google Scholar
Digital Library
- F. Cacheda, V. Carneiro, D. Fernández, and V. Formoso. 2011. Comparison of collaborative filtering algorithms: Limitations of current techniques and proposals for scalable, high-performance recommender systems. ACM Transactions on the Web 5, 1, 2. Google Scholar
Digital Library
- L. Candillier, F. Meyer, and F. Fessant. 2008. Designing specific weighted similarity measures to improve collaborative filtering systems. In Proceedings of the 8th Industrial Conference on Advances in Data Mining (ICDM’08). 242--255. Google Scholar
Digital Library
- C. Desrosiers and G. Karypis. 2011. A comprehensive survey of neighborhood-based recommendation methods. Recommender Systems Handbook 107--144.Google Scholar
- G. Guo, J. Zhang, and D. Thalmann. 2012. A simple but effective method to incorporate trusted neighbors in recommender systems. In Proceedings of the 20th International Conference on User Modeling, Adaptation and Personalization (UMAP’12). 114--125. Google Scholar
Digital Library
- G. Guo, J. Zhang, and N. Yorke-Smith. 2013. A novel Bayesian similarity measure for recommender systems. In Proceedings of the 23rd International Joint Conference on Artificial Intelligence (IJCAI’13). 2619--2625. Google Scholar
Digital Library
- Y. Huang, X. Chen, J. Zhang, D. Zeng, D. Zhang, and X. Ding. 2015. Single-trial {ERPs} denoising via collaborative filtering on {ERPs} images. Neurocomputing 149, Part B, 914--923. Google Scholar
Digital Library
- Y. Koren. 2010. Factor in the neighbors: Scalable and accurate collaborative filtering. ACM Transactions on Knowledge Discovery from Data 4, 1, 1:1--1:24. Google Scholar
Digital Library
- N. Lathia, S. Hailes, and L. Capra. 2007. Private distributed collaborative filtering using estimated concordance measures. In Proceedings of the 2007 ACM Conference on Recommender Systems (RecSys’07). 1--8. Google Scholar
Digital Library
- N. Lathia, S. Hailes, and L. Capra. 2008. The effect of correlation coefficients on communities of recommenders. In Proceedings of the 23rd Annual ACM Symposium on Applied Computing (SAC’08). 2000--2005. Google Scholar
Digital Library
- B. Li, X. Zhu, R. Li, C. Zhang, X. Xue, and X. Wu. 2011. Cross-domain collaborative filtering over time. In Proceedings of the 22nd International Joint Conference on Artificial Intelligence (IJCAI’11). 2293--2298. Google Scholar
Digital Library
- Dekang Lin. 1998. An information-theoretic definition of similarity. In Proceedings of the 15th International Conference on Machine Learning (ICML’98), Vol. 98. 296--304. Google Scholar
Digital Library
- X. Luo, Z. Ming, Z. You, S. Li, Y. Xia, and H. Leung. 2015. Improving network topology-based protein interactome mapping via collaborative filtering. Knowledge-Based Systems 90, 23--32. Google Scholar
Digital Library
- H. Ma, I. King, and M. R. Lyu. 2007. Effective missing data prediction for collaborative filtering. In Proceedings of the 30th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR’07). 39--46. Google Scholar
Digital Library
- H. Ma, D. Zhou, C. Liu, M. R. Lyu, and I. King. 2011. Recommender systems with social regularization. In Proceedings of the 4th ACM International Conference on Web Search and Data Mining (WSDM’11). 287--296. Google Scholar
Digital Library
- B. K. Mohan, B. J. Keller, and N. Ramakrishnan. 2007. Scouts, promoters, and connectors: The roles of ratings in nearest neighbor collaborative filtering. ACM Transactions on the Web 1, 2, 8. Google Scholar
Digital Library
- A. O’Hagan. 2004. Bayesian statistics: Principles and benefits. Frontis 3, 31--45.Google Scholar
- F. Ortega, J. L. Sánchez, J. Bobadilla, and A. Gutiérrez. 2013. Improving collaborative filtering-based recommender systems results using Pareto dominance. Information Sciences 239, 0, 50--61. Google Scholar
Digital Library
- W. Pan, Z. Liu, Z. Ming, H. Zhong, X. Wang, and C. Xu. 2015a. Compressed knowledge transfer via factorization machine for heterogeneous collaborative recommendation. Knowledge-Based Systems 85, 234--244. Google Scholar
Digital Library
- W. Pan, H. Zhong, C. Xu, and Z. Ming. 2015b. Adaptive Bayesian personalized ranking for heterogeneous implicit feedbacks. Knowledge-Based Systems 73, 173--180.Google Scholar
Digital Library
- Y. Ren, G. Li, J. Zhang, and W. Zhou. 2012. The efficient imputation method for neighborhood-based collaborative filtering. In Proceedings of the 21st ACM International Conference on Information and Knowledge Management. 684--693. Google Scholar
Digital Library
- S. J. Russell and P. Norvig. 2009. Artificial Intelligence: A Modern Approach (3rd ed.). Prentice Hall, Englewood Cliffs, NJ. Google Scholar
Digital Library
- A. Said, B. J. Jain, and S. Albayrak. 2012. Analyzing weighting schemes in collaborative filtering: Cold start, post cold start and power users. In Proceedings of the 27th Annual ACM Symposium on Applied Computing (SAC’12). 2035--2040. Google Scholar
Digital Library
- U. Shardanand and P. Maes. 1995. Social information filtering: Algorithms for automating “word of mouth.” In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (SIGCHI’95). 210--217. Google Scholar
Digital Library
- Y. Shi, M. Larson, and A. Hanjalic. 2009. Exploiting user similarity based on rated-item pools for improved user-based collaborative filtering. In Proceedings of the 2009 ACM Conference on Recommender Systems (RecSys’09). 125--132. Google Scholar
Digital Library
- Y. Shi, M. Larson, and A. Hanjalic. 2013. Mining contextual movie similarity with matrix factorization for context-aware recommendation. ACM Transactions on Intelligent Systems and Technology 4, 1, 16. Google Scholar
Digital Library
- J. Wang, J. Z. Huang, J. Guo, and Y. Lan. 2015. Recommending high-utility search engine queries via a query-recommending model. Neurocomputing. Google Scholar
Digital Library
Index Terms
A Novel Evidence-Based Bayesian Similarity Measure for Recommender Systems
Recommendations
A novel Bayesian similarity measure for recommender systems
IJCAI '13: Proceedings of the Twenty-Third international joint conference on Artificial IntelligenceCollaborative filtering, a widely-used user-centric recommendation technique, predicts an item's rating by aggregating its ratings from similar users. User similarity is usually calculated by cosine similarity or Pearson correlation coefficient. However,...
Frequency-based similarity measure for multimedia recommender systems
Personalized recommendation has become a pivotal aspect of online marketing and e-commerce as a means of overcoming the information overload problem. There are several recommendation techniques but collaborative recommendation is the most effective and ...
A sub-one quasi-norm-based similarity measure for collaborative filtering in recommender systems
AbstractCollaborative filtering (CF) is one of the most successful approaches for an online store to make personalized recommendations through its recommender systems. A neighborhood-based CF method makes recommendations to a target customer ...






Comments