A New Trust-Based Collaborative Filtering Measure Using Bhattacharyya Coefficient
With the rapid growth of network data and demands of users, the concept of AI in recommendation system has become a hot academic topic. However, in sparse data, it is difficult for the current user to obtain his efficient neighbors and for some cold-start users, it doesn’t do anything. Therefore, we constructed a new measure of trust between users for neighborhood based on Collaborative filtering(CF) which uses a pair of users common ratings and exploits Bhattacharyya similarity to finds relevance of each pair of rated items. We also have measured the validity of the proposed model through accuracy, recall rate and F1 measures. The results show that although some recall rates will be lost, the precision is greatly improved. Overall, it achieved good results.
KeywordsCollaborative filtering Trust method Bhattacharyya coefficient Sparsity problem
The paper is funded by the National Natural Science Foundation of Grant No. 61272036. Meanwhile, it is also funded by the Central University Fundamental Research Fund and the Key Discipline of Shanghai Second Polytechnic University. The grant numbers are NZ2013306 and XXKZD1604 respectively.
- 1.Park, S.T., Pennock, D.M.: Applying collaborative filtering techniques to movie search for better ranking and browsing. In: Proceedings of the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 550–559. ACM, New York (2007) https://doi.org/10.1145/1281192.1281252
- 5.Golbeck, J., Hendler, J.: FilmTrust: Movie recommendations using trust in web-based social networks. In: Proceedings of the IEEE Consumer Communications and Networking Conference, vol. 96, pp. 282–286(2006). https://doi.org/10.1109/CCNC.2006.1593032
- 6.Jamali, M., Ester, M.: TrustWalker: a random walk model for combining trust-based and item-based recommendation. In: Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 397–406. ACM, New York (2009). https://doi.org/10.1145/1557019.1557067
- 7.Jamali, M., Ester, M.: A matrix factorization technique with trust propagation for recommendation in social networks. In: Proceedings of the fourth ACM Conference on Recommender Systems, pp. 135–142. ACM, Barcelona (2010). https://doi.org/10.1145/1864708.1864736
- 8.Jin, J., Chen, Q.: A trust-based Top-K recommender system using social tagging network. In: 2012 9th International Conference on Fuzzy Systems and Knowledge Discovery, pp. 1270–1274. IEEE, Sichuan (2012). https://doi.org/10.1109/FSKD.2012.6234277
- 10.Forsati, R., Barjasteh, I., Masrour, F., Esfahanian, A.H., Radha, H.: PushTrust: an efficient recommendation algorithm by leveraging trust and distrust relations. In: Proceedings of the 9th ACM Conference on Recommender Systems, pp. 51–58. ACM, Vienna (2015). https://doi.org/10.1145/2792838.2800198
- 11.Reshma, M., Pillai, R.R.: Semantic based trust recommendation system for social networks using virtual groups. In: 2016 International Conference on Next Generation Intelligent Systems (ICNGIS), pp. 1–6. IEEE, Kottayam (2016). https://doi.org/10.1109/ICNGIS.2016.7854045
- 12.Patra, B.K., Launonen, R., Ollikainen, V., Nandi, S.: Exploiting Bhattacharyya similarity measure to diminish user cold-start problem in sparse data. In: Džeroski, S., Panov, P., Kocev, D., Todorovski, L. (eds.) DS 2014. LNCS (LNAI), vol. 8777, pp. 252–263. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-11812-3_22CrossRefGoogle Scholar