耦合社会信任信息的矩阵分解协同过滤模型
|
摘要
在过去的十年中,协同过滤(CF)推荐系统已经取得了巨大的成功。然而,用户-物品矩阵的稀疏性和冷启动问题仍然是一个挑战。在线社交网络的出现,为推荐系统提供了大量社交网络信任信息,从而为解决这一问题提供了契机。该文基于矩阵分解协同过滤方法,提出了一种集成用户信任信息的模型。该方法利用用户信任信息对用户隐因子进行修正,采用自编码器来提取用户和物品隐特征向量的初始化特征,并针对社交网络中的信任关系提出了信任群组的检测算法。大规模的真实数据集上进行的广泛的实验表明,该模型与相关算法对比,不但能有效缓解冷启动,而且取得了更好的推荐性能。
Collaborative filtering(CF) recommender system has been a most successful recommendation model in the past decade. However, the sparseness of user-item matrix and cold-tart problem still remain the challenges. The emergence of online social networking provides a great deal of social trust information for recommender systems, thus providing an opportunity to solve these problems. In this paper, based on matrix factorization collaborative filtering method, a model of integrating user trust information is proposed. This method uses trust information of users to amend the user latent factors and employs an auto-encoder to extract the initialization features of user and item latent feature vectors. And then a trust group detection algorithm is proposed for the trust relationship in the social network. Extensive experiments on real data sets show that the proposed model can not only effectively alleviate cold start, but also achieve better recommendation performance than the compared algorithms.
Collaborative filtering(CF) recommender system has been a most successful recommendation model in the past decade. However, the sparseness of user-item matrix and cold-tart problem still remain the challenges. The emergence of online social networking provides a great deal of social trust information for recommender systems, thus providing an opportunity to solve these problems. In this paper, based on matrix factorization collaborative filtering method, a model of integrating user trust information is proposed. This method uses trust information of users to amend the user latent factors and employs an auto-encoder to extract the initialization features of user and item latent feature vectors. And then a trust group detection algorithm is proposed for the trust relationship in the social network. Extensive experiments on real data sets show that the proposed model can not only effectively alleviate cold start, but also achieve better recommendation performance than the compared algorithms.
引文
[1]SARWAR B,KARYPIS G,KONSTAN J,et al.Item-based collaborative filtering recommendation algorithms[C]//Proceedings of the International Conference on World Wide Web.New York:ACM Press,2001:285-295.
[2]MA Hao,YANG Hai-xuan,LYU M R,et al.Sorec:Social recommendation using probabilistic matrix factorization[C]//Proceedings of the ACM Conference on Information and Knowledge Management.California:ACM Press,2008:265-275.
[3]YANG Bo,LEI Yu,LIU Da-yong,et al.Social collaborative filtering by trust[C]//Proceedings of the 23rd International Joint Conference on Artificial Intelligence.Beijing:AAAIPress,2013:2747-2753.
[4]YANG Bo,LEI Yu,LIU Da-Yong,et al.Social collaborative filtering by trust[J].IEEE Transactions.on Pattern Analysis and Machine Intelligence,2017,39(8):1633-1647.
[5]MA Hao,KING I,LYU M R.Learning to recommend with social trust ensemble[C]//Proceedings of the International ACM SIGIR Conference on Research and Development in Information Retrieval.Boston:ACM Press,2009:19-23.
[6]JAMALI M,ESTER M.A matrix factorization technique with trust propagation for recommendation in social networks[C]//Proceedings of the ACM Conference on Recommender Systems.Barcelona:ACM Press,2010:26-30.
[7]MA Hao,ZHOU Deng-yong,LIU Chao,et al.Recommender systems with social regularization[C]//ACMInternational Conference on Web Search and Data Mining.Hong Kong,China:ACM Press,2011:287-296.
[8]LI Hui,WU Ding-ming,Tang Wen-bin,et al.Overlapping community regularization for rating prediction in social recommender systems[C]//Proceedings of the 9th ACMConference on Recommender Systems.Vienna:ACM Press,2015:27-34.
[9]TANG Ji-liang,WANG Su-hang,HU Xia,et al.Recommendation with social dimensions[C]//Proceedings of the 30th AAAI Conference on Artificial Intelligence.Honolulu:AAAI Press,2016:251-257.
[10]SEDHAIN S,MENON A K,SANNER S,et al.Low-Rank linear cold-start recommendation from social data[C]//Proceedings of the 31th AAAI Conference on Artificial Intelligence.San Francisco:AAAI Press,2017:1502-1508.
[11]HIDASI B,TIKK D.Initializing matrix factorization methods on implicit feedback databases[J].Journal of Universal Computer Science,2013,19(12):1834-1853.
[12]REZAEI M,BOOSTANI R,REZAEI M.An efficient initialization method for nonnegative matrix factorization[J].Journal of Applied Sciences,2011,11(2):354-359.
[13]RIFAI S,VINCENT P,MULLER X,et al.Contractive auto-encoders:Explicit invariance during feature extraction[C]//Proceedings of the International Conference on Machine Learning.Washington:IMLS,2011:833-840.
[14]FORTUNATO S.Community detection in graphs[J].Physics Reports,2009,486(3):75-174.
[15]ZHOU Zhen,WANG Wei,WANG Liang.Community detection based on an improved modularity[M].Berlin:Springer Berlin Heidelberg,2012:638-645.
[16]BLONDEL V D,GUILAUME J L,LAMBIOTTE R,et al.Fast unfolding of communities in large networks[J].Journal of Statistical Mechanics,2008(10):155-168.
[17]ZHANG He,YANG Zhi-rong,OJA E.Improving cluster analysis by co-initializations[J].Pattern Recognition Letters,2014,45(11):71-77.
[18]MASSA P,AVESANI P.Trust metrics in recommender systems[M].London:Springer London,2009:259-285.
[19]MASSA P,AVESANI P.Trust-aware recommender systems[C]//Proceedings of the ACM Conference on Recommender Systems.Minneapolis:ACM Press,2007:17-24.
[20]KANNAN R,ISHTEVA M,PARK H.Bounded matrix factorization for recommender system[J].Knowledge&Information Systems,2014,39(3):491-511.
[21]GU Quan-quan,ZHOU Jie,DING C.Collaborative filtering:Weighted nonnegative matrix factorization incorporating user and item graphs[C]//Proceedings of the SIAM International Conference on Data Mining.Ohio:SIAM Press,2010:199-210.
[2]MA Hao,YANG Hai-xuan,LYU M R,et al.Sorec:Social recommendation using probabilistic matrix factorization[C]//Proceedings of the ACM Conference on Information and Knowledge Management.California:ACM Press,2008:265-275.
[3]YANG Bo,LEI Yu,LIU Da-yong,et al.Social collaborative filtering by trust[C]//Proceedings of the 23rd International Joint Conference on Artificial Intelligence.Beijing:AAAIPress,2013:2747-2753.
[4]YANG Bo,LEI Yu,LIU Da-Yong,et al.Social collaborative filtering by trust[J].IEEE Transactions.on Pattern Analysis and Machine Intelligence,2017,39(8):1633-1647.
[5]MA Hao,KING I,LYU M R.Learning to recommend with social trust ensemble[C]//Proceedings of the International ACM SIGIR Conference on Research and Development in Information Retrieval.Boston:ACM Press,2009:19-23.
[6]JAMALI M,ESTER M.A matrix factorization technique with trust propagation for recommendation in social networks[C]//Proceedings of the ACM Conference on Recommender Systems.Barcelona:ACM Press,2010:26-30.
[7]MA Hao,ZHOU Deng-yong,LIU Chao,et al.Recommender systems with social regularization[C]//ACMInternational Conference on Web Search and Data Mining.Hong Kong,China:ACM Press,2011:287-296.
[8]LI Hui,WU Ding-ming,Tang Wen-bin,et al.Overlapping community regularization for rating prediction in social recommender systems[C]//Proceedings of the 9th ACMConference on Recommender Systems.Vienna:ACM Press,2015:27-34.
[9]TANG Ji-liang,WANG Su-hang,HU Xia,et al.Recommendation with social dimensions[C]//Proceedings of the 30th AAAI Conference on Artificial Intelligence.Honolulu:AAAI Press,2016:251-257.
[10]SEDHAIN S,MENON A K,SANNER S,et al.Low-Rank linear cold-start recommendation from social data[C]//Proceedings of the 31th AAAI Conference on Artificial Intelligence.San Francisco:AAAI Press,2017:1502-1508.
[11]HIDASI B,TIKK D.Initializing matrix factorization methods on implicit feedback databases[J].Journal of Universal Computer Science,2013,19(12):1834-1853.
[12]REZAEI M,BOOSTANI R,REZAEI M.An efficient initialization method for nonnegative matrix factorization[J].Journal of Applied Sciences,2011,11(2):354-359.
[13]RIFAI S,VINCENT P,MULLER X,et al.Contractive auto-encoders:Explicit invariance during feature extraction[C]//Proceedings of the International Conference on Machine Learning.Washington:IMLS,2011:833-840.
[14]FORTUNATO S.Community detection in graphs[J].Physics Reports,2009,486(3):75-174.
[15]ZHOU Zhen,WANG Wei,WANG Liang.Community detection based on an improved modularity[M].Berlin:Springer Berlin Heidelberg,2012:638-645.
[16]BLONDEL V D,GUILAUME J L,LAMBIOTTE R,et al.Fast unfolding of communities in large networks[J].Journal of Statistical Mechanics,2008(10):155-168.
[17]ZHANG He,YANG Zhi-rong,OJA E.Improving cluster analysis by co-initializations[J].Pattern Recognition Letters,2014,45(11):71-77.
[18]MASSA P,AVESANI P.Trust metrics in recommender systems[M].London:Springer London,2009:259-285.
[19]MASSA P,AVESANI P.Trust-aware recommender systems[C]//Proceedings of the ACM Conference on Recommender Systems.Minneapolis:ACM Press,2007:17-24.
[20]KANNAN R,ISHTEVA M,PARK H.Bounded matrix factorization for recommender system[J].Knowledge&Information Systems,2014,39(3):491-511.
[21]GU Quan-quan,ZHOU Jie,DING C.Collaborative filtering:Weighted nonnegative matrix factorization incorporating user and item graphs[C]//Proceedings of the SIAM International Conference on Data Mining.Ohio:SIAM Press,2010:199-210.