利用Tri-training算法解决推荐系统冷启动问题

doi:10.11896/j.issn.1002-137X.2016.12.019

摘要/Abstract

摘要： 随着社交网络的发展,推荐系统日趋重要,而冷启动问题是推荐系统中的关键问题。设计了一种基于上下文的半监督学习框架TSEL,对矩阵分解模型SVD进行扩充以支持更多形式的上下文信息,利用Tri-training框架训练各个模型。与其他解决推荐系统冷启动问题的半监督方法(如Co- training)相比,该方法有着更好的效果。Tri-training框架能够更加方便地引入更多推荐模型,具有更好的可扩展性。将Tri-training框架加以扩展,提出了基于用户活跃度生成无标记教学集合的算法和更加丰富的对矩阵分解模型扩充的形式。在真实数据集MovieLens上进行验证,获得了更好的实验效果。

关键词: 推荐系统,机器学习,Tri-training

Abstract: With the development of social network,recommender system is becoming more and more important.Cold start is one of the most important problems in recommender system.A context-based semi-supervised learning framework TSEL was designed.We expanded matrix factorization model SVD to support more kinds of context information,and used Tri-training framework to train individual models.Compared with other methods which solve cold start problems in recommender system (e.g.Co-training),our algorithm has better performance.Tri-training framework can incorporate more recommender models and has good expansibility.We expanded Tri-training framework,and proposed a user activeness-based unlabeled teaching set generating algorithm.We proposed more kinds of models which expand the matrix factorization.We evaluated our algorithm on real world dataset,i.e.MovieLens,and got better performance.

Key words: Recommender system,Machine learning,Tri-training

张栩晨. 利用Tri-training算法解决推荐系统冷启动问题[J]. 计算机科学, 2016, 43(12): 108-114. https://doi.org/10.11896/j.issn.1002-137X.2016.12.019

ZHANG Xu-chen. Utilizing Tri-training Algorithm to Solve Cold Start Problem in Recommender System[J]. Computer Science, 2016, 43(12): 108-114. https://doi.org/10.11896/j.issn.1002-137X.2016.12.019

参考文献

[1] Ma H W,Zhang G W,Li P,et al.Introduction to Callaborative Filtering in Recommender System[J].Journal of Chinese Computer Systems,2009,30(7):1282-1288(in Chinese) 马宏伟,张光卫,李鹏,等.协同过滤推荐算法综述[J].小型微型计算机系统,2009,30(7):1282-1288
[2] Schein,Andrew I,Popescul,et al.Methods and Metrics for Cold-Start Recommendations[C]∥ Proceedings of the 25th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval.2002:253-260
[3] Lin J,Sugiyama K,Kan M Y,et al.Addressing cold-start in app recommendation:latent user models constructed from twitter followers[C]∥Proceedings of the 36th International ACM SIGIR Conference on Research and Development in Information Retrieval.ACM,2013:283-292
[4] Sarwar,Badrul M,Konstan,et al.Using Filtering Agents to Improve Prediction Quality in the GroupLens Research Collaborative Filtering System[C]∥Acm Conference on Computer Supported Cooperative Work.1999:345-354
[5] Deshpande M,Karypis G.Item-Based Top-N RecommendationAlgorithms[J].ACM Transactions on Information Systems,2004,22(1):143-177
[6] Degemmis M,Lops P,Semeraro G.A content-collaborative re-commender that exploits WordNet-based user profiles for neighborhood formation[J].User Modeling and User-Adapted Inte-raction,2007,17(3):217-255
[7] Guo G.Improving the performance of recommender systems by alleviating the data sparsity and cold start problems[C]∥Proceedings of the Twenty-Third International Joint Conference on Artificial Intelligence.AAAI Press,2013:3217-3218
[8] Tang J,Wu S,Sun J,et al.Cross-domain collaboration recommendation[C]∥Kdd.2012:1285-1293
[9] Takács G,Pilászy I,Németh B,et al.Scalable Collaborative Filtering Approaches for Large Recommender Systems[J].Journal of Machine Learning Research,2009,10(2):623-656
[10] Koenigstein N,Dror G,Koren Y.Yahoo! music recommenda-tions:modeling music ratings with temporal dynamics and item taxonomy[C]∥Fifth ACM Conference on Recommender Systems.ACM,2011:165-172
[11] Park S T,Chu W.Pairwise preference regression for cold-start recommendation[C]∥RecSys 2009.2009:21-28
[12] Zhou K,Yang S H,Zha H.Functional Matrix Factorizations for Cold-Start Recommendation[C]∥Proceedings of the 34th International ACM SIGIR Conference on Research and Development in Information Retrieval.ACM,2011:315-324
[13] Sun Ming-xuan,Li Fu-xin,Lee J ,et al.Learning multiple-question decision trees for cold-start recommendation[C]∥Procee-dings of the sixth ACM International Conference on Web Search and Data Mining.ACM,2013:445-454
[14] Zhou Z H,Li M.Semisupervised Regression with Cotraining-Style Algorithms[J].IEEE Transactions on Knowledge & Data Engineering,2007,19(11):1479-1493
[15] Koren,Yehuda.Factorization meets the neighborhood:a multi-faceted collaborative filtering model[C]∥AcmSigkdd International Conference on Knowledge Discovery & Data Mining.2008:426-434
[16] Breiman L.Bagging Predictors[J].Machine Learning,1996,24(2):123-140
[17] Zhou Z H,Li M.Semi-Supervised Regression with Tri-training[C]∥Proceedings of the 19th International Joint Conference on Artificial Intelligence.Morgan Kaufmann Publishers Inc.,2005:908-916
[18] Rokach L.Ensemble-based classifiers[J].Artificial Intelligence Review,2010,33(1):1-39
[19] Zhang M L,Zhou Z H.Exploiting unlabeled data to enhance ensemble diversity[J].Data Mining & Knowledge Discovery,2013,26(1):98-129
[20] Zhang M,Tang J,Zhang X,et al.Addressing cold start in re-commender systems:a semi-supervised Co-training algorithm[C]∥International Conference on Research on Development in Information Retrieval.2014
[21] Bellaachia A,Alathel D.Improving the Recommendation Accuracy for Cold Start Users in Trust-Based Recommender Systems[J].International Journal of Computer and Communication Engineering,2016,5(3):206
[22] Li L,Tang X.A Solution to the Cold- Start Problem in Recommender Systems Based on Social Choice Theory[M]∥Intelligent and Evolutionary Systems.Springer International Publi-shing,2016:267-279

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed