Abstract: Implicit feedback, which indirectly reflects opinion through user behaviors, has gained increasing attention in recommender system communities due to its accessibility and richness in real-world applications. A major way of exploiting implicit feedback is to treat the data as an indication of positive and negative preferences associated with vastly varying confidence levels. Such algorithms assume that the numerical value of implicit feedback, such as time of watching, indicates confidence, rather than degree of preference, and a larger value indicates a higher confidence, although this works only when just one type of implicit feedback is available. However, in real-world applications, there are usually various types of implicit feedback, which can be referred to as heterogeneous implicit feedback. Existing methods cannot efficiently infer confidence levels from heterogeneous implicit feedback. In this paper, we propose a novel confidence estimation approach to infer the confidence level of user preference based on heterogeneous implicit feedback. Then we apply the inferred confidence to both point-wise and pair-wise matrix factorization models, and propose a more generic strategy to select effective training samples for pair-wise methods. Experiments on real-world e-commerce datasets from Tmall.com show that our methods outperform the state-of-the-art approaches, considering several commonly used ranking-oriented evaluation criteria.

针对异构隐式反馈的置信度估计方法

[1]Bobadilla, J., Ortega, F., Hernando, A., et al., 2013. Recommender systems survey. Knowl.-Based Syst., 46:109-132.

[2]Freedman, D.A., 2009. Statistical Models: Theory and Practice (2nd Editon). Cambridge University Press, Cambridge.

[3]Friedman, J.H., 2001. Greedy function approximation: a gradient boosting machine. Ann. Stat., 29(5):1189-1232.

[4]Friedman, J.H., 2002. Stochastic gradient boosting. Comput. Stat. Data Anal., 38(4):367-378.

[5]Hu, Y.F., Koren, Y., Volinsky, C., 2008. Collaborative filtering for implicit feedback datasets. Proc. 8th IEEE Int. Conf. on Data Mining, p.263-272.

[6]Koren, Y., 2008. Factorization meets the neighborhood: a multifaceted collaborative filtering model. Proc. 14th ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, p.426-434.

[7]Koren, Y., 2010. Factor in the neighbors: scalable and accurate collaborative filtering. ACM Trans. Knowl. Discov. Data, 4(1), Article 1.

[8]Lee, T.Q., Park, Y., Park, Y.T., 2008. A time-based approach to effective recommender systems using implicit feedback. Expert Syst. Appl., 34(4):3055-3062.

[9]Liaw, A., Wiener, M., 2002. Classification and regression by randomForest. R News, 2(3):18-22.

[10]Pan, R., Zhou, Y.H., Cao, B., et al., 2008. One-class collaborative filtering. Proc. 8th IEEE Int. Conf. on Data Mining, p.502-511.

[11]Pan, W.K., Zhong, H., Xu, C.F., et al., 2015. Adaptive Bayesian personalized ranking for heterogeneous implicit feedbacks. Knowl.-Based Syst., 73:173-180.

[12]Pan, W.K., Liu, M.S., Ming, Z., 2016. Transfer learning for heterogeneous one-class collaborative filtering. IEEE Intell. Syst., 31(4):43-49.

[13]Park, D.H., Kim, H.K., Choi, I.Y., et al., 2012. A literature review and classification of recommender systems research. Expert Syst. Appl., 39(11):10059-10072.

[14]Rendle, S., Freudenthaler, C., Gantner, Z., et al., 2009. BPR: Bayesian personalized ranking from implicit feedback. Proc. 25th Conf. on Uncertainty in Artificial Intelligence, p.452-461.

[15]Ricci, F., Rokach, L., Shapira, B., et al., 2011. Recommender Systems Handbook. Springer, Boston, MA, US.

[16]Shi, Y., Larson, M., Hanjalic, A., 2014. Collaborative filtering beyond the user-item matrix: a survey of the state of the art and future challenges. ACM Comput. Surv., 47(1):1-45.

[17]Tuzhilin, A., Adomavicius, G., 2005. Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions. IEEE Trans. Knowl. Data Eng., 17(6):734-749.

[18]Volkovs, M., Yu, G.W., 2015. Effective latent models for binary feedback in recommender systems. Proc. 38th Int.

[19]ACM SIGIR Conf. on Research and Development in Information Retrieval, p.313-322.

[20]Wang, J., Lin, L.F., Zhang, H., et al., 2016. Confidence-learning based collaborative filtering with heterogeneous implicit feedbacks. Proc. 18th Asia-Pacific Web Conf., p.444-455.

[21]Wang, S., Zhou, X.B., Wang, Z.Q., et al., 2012. Please spread: recommending tweets for retweeting with implicit feedback. Proc. Workshop on Data-Driven User Behavioral Modelling and Mining from Social Media, p.19-22.