一种面向多源异构数据的协同过滤推荐算法
|
摘要
随着电子商务网站的快速发展,数据特征和现实需求均发生了较大变化.以大规模、多源性、异构性为主要特征的数据发挥着更加重要的作用.然而,电子商务系统中数据所具有的特性使得大多数协同过滤方法较难直接用于物品推荐.如何整合多源异构数据来实现数据的价值最大化是当前推荐系统亟待解决的问题.针对这一问题,首先分析了多源异构数据中各类数据的特点,并根据各自特点为其设计了不同的建模方式.其次,提出一种新颖的推荐模型用于评分预测任务,它通过融合多关系数据和视觉信息来缓解数据稀疏问题.最后,设计了一种高效的算法MSRA(multi-source heterogeneous information based recommendation algorithm)用于求解所提模型的参数.在多个亚马逊数据集上的实验结果表明:1)面向多源异构数据的推荐算法其性能明显优于当前主流协同过滤算法; 2)该算法不仅可以有效缓解物品的冷启动问题,而且能够更好地预测不同类型物品的实际评分.
With the rapid development of electronic e-commerce sites, data characteristics and realistic demands have changed. The data, which has main characteristics of large-scale, multi-source and heterogeneous, is playing an important role. However, these unique characteristics of electronic e-commerce systems make most of existing collaborative filtering methods difficult to be adapted for product recommendation. The immediate problem to be solved is how to integrate multi-source heterogeneous data to achieve the maximum value of big data. In this paper, we first analyze the characteristics of various data among different information sources, and design different modeling solutions. Then, we propose a novel recommendation model for the task of rating prediction, which makes it possible to mitigate the sparsity problem via seamlessly integrating multi-relational data and visual contents. Finally, we devise a computationally efficient learning algorithm named MSRA(multi-source heterogeneous information based recommendation algorithm), to optimize the proposed model. To verify the effectiveness of our proposed model, we conduct extensive experiments on a wide spectrum of large-scale Amazon datasets. Experimental results demonstrate that 1)the designed algorithm consistently and significantly outperforms several state-of-the-art collaborative filtering algorithms, and 2)our algorithm is capable of alleviating the item cold-start problem and helping obtain more accurate results of various items.
With the rapid development of electronic e-commerce sites, data characteristics and realistic demands have changed. The data, which has main characteristics of large-scale, multi-source and heterogeneous, is playing an important role. However, these unique characteristics of electronic e-commerce systems make most of existing collaborative filtering methods difficult to be adapted for product recommendation. The immediate problem to be solved is how to integrate multi-source heterogeneous data to achieve the maximum value of big data. In this paper, we first analyze the characteristics of various data among different information sources, and design different modeling solutions. Then, we propose a novel recommendation model for the task of rating prediction, which makes it possible to mitigate the sparsity problem via seamlessly integrating multi-relational data and visual contents. Finally, we devise a computationally efficient learning algorithm named MSRA(multi-source heterogeneous information based recommendation algorithm), to optimize the proposed model. To verify the effectiveness of our proposed model, we conduct extensive experiments on a wide spectrum of large-scale Amazon datasets. Experimental results demonstrate that 1)the designed algorithm consistently and significantly outperforms several state-of-the-art collaborative filtering algorithms, and 2)our algorithm is capable of alleviating the item cold-start problem and helping obtain more accurate results of various items.
引文
[1]Adomavicius G,Tuzhilin A.Toward the next generation of recommender systems:A survey of the state-of-the-art and possible extensions[J].IEEE Transactions on Knowledge and Data Engineering,2005,17(6):734- 749
[2]Sarwar B,Karypis G,Konstan J,et al.Item-based collaborative filtering recommendation algorithms[C] //Proc of the 10th Int Conf on World Wide Web (WWW’01).New York:ACM,2001:285- 295
[3]Smith B,Linden G.Two decades of recommender systems at amazon.com[J].IEEE Internet Computing,2017,21(3):12- 18
[4]Weimer M,Karatzoglou A,Viet Le Q,et al.CoFIRank:Maximum margin matrix factorization for collaborative ranking[C] //Proc of the 22nd Annual Conf on Neural Information Processing Systems (NIPS’08).New York:Curran Associates Inc,2008:1593- 1600
[5]Mnih A,Salakhutdinov R.Probabilistic matrix factorization[C] //Proc of the 21st Annual Conf on Neural Information Processing Systems (NIPS’07).New York:Curran Associates Inc,2007:1257- 1264
[6]Koren Y,Bell R,Volinsky C.Matrix factorization techniques for recommender systems[J].IEEE Computer,2009,42(8):30- 37
[7]Koren Y.Factorization meets the neighborhood:A multifaceted collaborative filtering model[C] //Proc of the 14th ACM SIGKDD Int Conf on Knowledge Discovery and Data Mining (KDD’08).New York:ACM,2008:426- 434
[8]Wu Le,Liu Qi,Chen Enhong,et al.Relevance meets coverage:A unified framework to generate diversified recommendations[J].ACM Transactions on Intelligent Systems and Technology,2016,7(3):39:1- 39:30
[9]Liu Qi,Chen Enhong,Xiong Hui,et al.Enhancing collaborative filtering by user interest expansion via personalized ranking[J].IEEE Transactions on Systems,Man,and Cybernetics:Part B (Cybernetics),2012,42(1):218- 233
[10]Zhang Kaihan,Liang Jiye,Zhao Xingwang,et al.A collaborative filtering recommendation algorithm based on information of community experts[J].Journal of Computer Research and Development,2018,55(5):968- 976 (in Chinese)(张凯涵,梁吉业,赵兴旺,等.一种基于社区专家信息的协同过滤推荐算法[J].计算机研究与发展,2018,55(5):968- 976)
[11]Yu Yonghong,Gao Yang,Wang Hao,et al.Integrating user social status and matrix factorization for item recommendation[J].Journal of Computer Research and Development,2018,55(1):113- 124 (in Chinese)(余永红,高阳,王皓,等.融合用户社会地位和矩阵分解的推荐算法[J].计算机研究与发展,2018,55(1):113- 124)
[12]Guo Lei,Ma Jun,Chen Zhumin.Trust strength aware social recommendation method[J].Journal of Computer Research and Development,2013,50(9):1805- 1813 (in Chinese)(郭磊,马军,陈竹敏.一种信任关系强度敏感的社会化推荐算法[J].计算机研究与发展,2013,50(9):1805- 1813)
[13]Ma Hao,Zhou Dengyong,Liu Chao,et al.Recommender systems with social regularization[C] //Proc of the 4th ACM Conf on Web Search and Data Mining (WSDM’11).New York:ACM,2011:287- 296
[14]Jamali M,Ester M.A matrix factorization technique with trust propagation for recommendation in social networks[C] //Proc of the 4th ACM Conf on Recommender Systems (RecSys’10).New York:ACM,2010:135- 142
[15]Guo Guibing,Zhang Jie,Yorke-Smith N.TrustSVD:Collaborative filtering with both the explicit and implicit influence of user trust and of item ratings[C] //Proc of the 29th AAAI Conf on Artificial Intelligence (AAAI’15).Menlo Park,CA:AAAI,2015:123- 129
[16]Li Xin,Liu Guiquan,Li Lin,et al.Circle-based and social connection embedded recommendation in LBSN[J].Journal of Computer Research and Development,2017,54(2):394- 404 (in Chinese)(李鑫,刘贵全,李琳,等.LBSN上基于兴趣圈中社会关系挖掘的推荐算法[J].计算机研究与发展,2017,54(2):394- 404)
[17]Sun Guangfu,Wu Le,Liu Qi,et al.Recommendations based on collaborative filtering by exploiting sequential behaviors[J].Journal of Software,2013,24(11):2721- 2733 (in Chinese)(孙光福,吴乐,刘淇,等.基于时序行为的协同过滤推荐算法[J].软件学报,2013,24(11):2721- 2733)
[18]Wu Le,Chen Enhong,Liu Qi,et al.Leveraging tagging for neighborhood-aware probabilistic matrix factorization[C] //Proc of the 21st ACM Conf on Information and Knowledge Management (CIKM’12).New York:ACM,2012:1854- 1858
[19]Gu Quanquan,Zhou Jie,Ding Chris.Collaborative filtering:Weighted nonnegative matrix factorization incorporating user and item graphs[C] //Proc of the 2010 SIAM Int Conf on Data Mining (SDM’10).Philadelphia,PA:SIAM,2010:199- 210
[20]Krizhevsky A,Sutskever I,Hinton G E.Imagenet classification with deep convolutional neural networks[C] //Proc of the 21st Annual Conf on Neural Information Processing Systems (NIPS’12).New York:Curran Associates Inc,2012:1097- 1105
[21]Graves A,Mohamed A,Hinton G.Speech recognition with deep recurrent neural networks[C] //Proc of the 2013 IEEE Int Conf on Acoustics,Speech and Signal Processing (ICASSP’13).Piscataway,NJ:IEEE,2013:6645- 6649
[22]He Ruining,McAuley J.VBPR:Visual Bayesian personalized ranking from implicit feedback[C] //Proc of the 30th AAAI Conf on Artificial Intelligence (AAAI’16).Menlo Park,CA:AAAI,2016:144- 150
[23]Jia Yangqing,Shelhamer E,Donahue J,et al.Caffe:Convolutional architecture for fast feature embedding[C] //Proc of the 22nd ACM Int Conf on Multimedia (MM’14).New York:ACM,2014:675- 678
[24]Rendle S,Freudenthaler C,Gantner Z,et al.BPR:Bayesian personalized ranking from implicit feedback[C] //Proc of the 25th Conf on Uncertainty in Artificial Intelligence (UAI’09).Arlington,Virginia:AUAI,2009:452- 461
[25]Roy S,Guntuku S C.Latent factor representations for cold-start video recommendation[C] //Proc of the 10th ACM Conf on Recommender Systems (RecSys’16).New York:ACM,2016:99- 106
[26]Wang Suhang,Wang Yilin,Tang Jiliang,et al.What your images reveal:Exploiting visual contents for point-of-interest recommendation[C] //Proc of the 26th Int Conf on World Wide Web (WWW’17).New York:ACM,2017:391- 400
[27]Wang Zihan,Jiang Ziheng,Ren Zhaochun,et al.A path-constrained framework for discriminating substitutable and complementary products in e-commerce[C] //Proc of the 8th ACM Conf on Web Search and Data Mining (WSDM’18).New York:ACM,2018:619- 627
[28]Hu Longke,Sun Aixin,Liu Yong.Your neighbors affect your ratings:On geographical neighborhood influence to rating prediction[C] //Proc of the 37th Int ACM SIGIR Conf on Research and Development in Information Retrieval (SIGIR’14).New York:ACM,2014:345- 354
[29]Singh A P,Gordon G J.Relational learning via collective matrix factorization[C] //Proc of the 14th ACM SIGKDD Int Conf on Knowledge Discovery and Data Mining (KDD’08).New York:ACM,2008:650- 658
[30]Meng Xiaofeng,Du Zhijuan.Research on the big data fusion:Issues and challenges[J].Journal of Computer Research and Development,2016,53(2):231- 246 (in Chinese)(孟小峰,杜治娟.大数据融合研究:问题与挑战[J].计算机研究与发展,2016,53(2):231- 246)
[31]Huang Shanshan,Wang Shuaiqiang,Liu Tieyan,et al.Listwise collaborative filtering[C] //Proc of the 38th Int ACM SIGIR Conf on Research and Development in Information Retrieval (SIGIR’15).New York:ACM,2015:343- 352
[32]Ma Hao.An experimental study on implicit social recommendation[C] //Proc of the 36th Int ACM SIGIR Conf on Research and Development in Information Retrieval (SIGIR’13).New York:ACM,2013:73- 82
[33]Fazeli S,Loni B,Bellogin A,et al.Implicit vs explicit trust in social matrix factorization[C] //Proc of the 8th ACM Conf on Recommender Systems (RecSys’14).New York:ACM,2014:317- 320
[34]Park C,Kim D,Oh J,et al.Do “also-viewed” products help user rating prediction?[C] //Proc of the 26th Int Conf on World Wide Web (WWW’17).New York:ACM,2017:1113- 1122
① http://www.librec.net
[2]Sarwar B,Karypis G,Konstan J,et al.Item-based collaborative filtering recommendation algorithms[C] //Proc of the 10th Int Conf on World Wide Web (WWW’01).New York:ACM,2001:285- 295
[3]Smith B,Linden G.Two decades of recommender systems at amazon.com[J].IEEE Internet Computing,2017,21(3):12- 18
[4]Weimer M,Karatzoglou A,Viet Le Q,et al.CoFIRank:Maximum margin matrix factorization for collaborative ranking[C] //Proc of the 22nd Annual Conf on Neural Information Processing Systems (NIPS’08).New York:Curran Associates Inc,2008:1593- 1600
[5]Mnih A,Salakhutdinov R.Probabilistic matrix factorization[C] //Proc of the 21st Annual Conf on Neural Information Processing Systems (NIPS’07).New York:Curran Associates Inc,2007:1257- 1264
[6]Koren Y,Bell R,Volinsky C.Matrix factorization techniques for recommender systems[J].IEEE Computer,2009,42(8):30- 37
[7]Koren Y.Factorization meets the neighborhood:A multifaceted collaborative filtering model[C] //Proc of the 14th ACM SIGKDD Int Conf on Knowledge Discovery and Data Mining (KDD’08).New York:ACM,2008:426- 434
[8]Wu Le,Liu Qi,Chen Enhong,et al.Relevance meets coverage:A unified framework to generate diversified recommendations[J].ACM Transactions on Intelligent Systems and Technology,2016,7(3):39:1- 39:30
[9]Liu Qi,Chen Enhong,Xiong Hui,et al.Enhancing collaborative filtering by user interest expansion via personalized ranking[J].IEEE Transactions on Systems,Man,and Cybernetics:Part B (Cybernetics),2012,42(1):218- 233
[10]Zhang Kaihan,Liang Jiye,Zhao Xingwang,et al.A collaborative filtering recommendation algorithm based on information of community experts[J].Journal of Computer Research and Development,2018,55(5):968- 976 (in Chinese)(张凯涵,梁吉业,赵兴旺,等.一种基于社区专家信息的协同过滤推荐算法[J].计算机研究与发展,2018,55(5):968- 976)
[11]Yu Yonghong,Gao Yang,Wang Hao,et al.Integrating user social status and matrix factorization for item recommendation[J].Journal of Computer Research and Development,2018,55(1):113- 124 (in Chinese)(余永红,高阳,王皓,等.融合用户社会地位和矩阵分解的推荐算法[J].计算机研究与发展,2018,55(1):113- 124)
[12]Guo Lei,Ma Jun,Chen Zhumin.Trust strength aware social recommendation method[J].Journal of Computer Research and Development,2013,50(9):1805- 1813 (in Chinese)(郭磊,马军,陈竹敏.一种信任关系强度敏感的社会化推荐算法[J].计算机研究与发展,2013,50(9):1805- 1813)
[13]Ma Hao,Zhou Dengyong,Liu Chao,et al.Recommender systems with social regularization[C] //Proc of the 4th ACM Conf on Web Search and Data Mining (WSDM’11).New York:ACM,2011:287- 296
[14]Jamali M,Ester M.A matrix factorization technique with trust propagation for recommendation in social networks[C] //Proc of the 4th ACM Conf on Recommender Systems (RecSys’10).New York:ACM,2010:135- 142
[15]Guo Guibing,Zhang Jie,Yorke-Smith N.TrustSVD:Collaborative filtering with both the explicit and implicit influence of user trust and of item ratings[C] //Proc of the 29th AAAI Conf on Artificial Intelligence (AAAI’15).Menlo Park,CA:AAAI,2015:123- 129
[16]Li Xin,Liu Guiquan,Li Lin,et al.Circle-based and social connection embedded recommendation in LBSN[J].Journal of Computer Research and Development,2017,54(2):394- 404 (in Chinese)(李鑫,刘贵全,李琳,等.LBSN上基于兴趣圈中社会关系挖掘的推荐算法[J].计算机研究与发展,2017,54(2):394- 404)
[17]Sun Guangfu,Wu Le,Liu Qi,et al.Recommendations based on collaborative filtering by exploiting sequential behaviors[J].Journal of Software,2013,24(11):2721- 2733 (in Chinese)(孙光福,吴乐,刘淇,等.基于时序行为的协同过滤推荐算法[J].软件学报,2013,24(11):2721- 2733)
[18]Wu Le,Chen Enhong,Liu Qi,et al.Leveraging tagging for neighborhood-aware probabilistic matrix factorization[C] //Proc of the 21st ACM Conf on Information and Knowledge Management (CIKM’12).New York:ACM,2012:1854- 1858
[19]Gu Quanquan,Zhou Jie,Ding Chris.Collaborative filtering:Weighted nonnegative matrix factorization incorporating user and item graphs[C] //Proc of the 2010 SIAM Int Conf on Data Mining (SDM’10).Philadelphia,PA:SIAM,2010:199- 210
[20]Krizhevsky A,Sutskever I,Hinton G E.Imagenet classification with deep convolutional neural networks[C] //Proc of the 21st Annual Conf on Neural Information Processing Systems (NIPS’12).New York:Curran Associates Inc,2012:1097- 1105
[21]Graves A,Mohamed A,Hinton G.Speech recognition with deep recurrent neural networks[C] //Proc of the 2013 IEEE Int Conf on Acoustics,Speech and Signal Processing (ICASSP’13).Piscataway,NJ:IEEE,2013:6645- 6649
[22]He Ruining,McAuley J.VBPR:Visual Bayesian personalized ranking from implicit feedback[C] //Proc of the 30th AAAI Conf on Artificial Intelligence (AAAI’16).Menlo Park,CA:AAAI,2016:144- 150
[23]Jia Yangqing,Shelhamer E,Donahue J,et al.Caffe:Convolutional architecture for fast feature embedding[C] //Proc of the 22nd ACM Int Conf on Multimedia (MM’14).New York:ACM,2014:675- 678
[24]Rendle S,Freudenthaler C,Gantner Z,et al.BPR:Bayesian personalized ranking from implicit feedback[C] //Proc of the 25th Conf on Uncertainty in Artificial Intelligence (UAI’09).Arlington,Virginia:AUAI,2009:452- 461
[25]Roy S,Guntuku S C.Latent factor representations for cold-start video recommendation[C] //Proc of the 10th ACM Conf on Recommender Systems (RecSys’16).New York:ACM,2016:99- 106
[26]Wang Suhang,Wang Yilin,Tang Jiliang,et al.What your images reveal:Exploiting visual contents for point-of-interest recommendation[C] //Proc of the 26th Int Conf on World Wide Web (WWW’17).New York:ACM,2017:391- 400
[27]Wang Zihan,Jiang Ziheng,Ren Zhaochun,et al.A path-constrained framework for discriminating substitutable and complementary products in e-commerce[C] //Proc of the 8th ACM Conf on Web Search and Data Mining (WSDM’18).New York:ACM,2018:619- 627
[28]Hu Longke,Sun Aixin,Liu Yong.Your neighbors affect your ratings:On geographical neighborhood influence to rating prediction[C] //Proc of the 37th Int ACM SIGIR Conf on Research and Development in Information Retrieval (SIGIR’14).New York:ACM,2014:345- 354
[29]Singh A P,Gordon G J.Relational learning via collective matrix factorization[C] //Proc of the 14th ACM SIGKDD Int Conf on Knowledge Discovery and Data Mining (KDD’08).New York:ACM,2008:650- 658
[30]Meng Xiaofeng,Du Zhijuan.Research on the big data fusion:Issues and challenges[J].Journal of Computer Research and Development,2016,53(2):231- 246 (in Chinese)(孟小峰,杜治娟.大数据融合研究:问题与挑战[J].计算机研究与发展,2016,53(2):231- 246)
[31]Huang Shanshan,Wang Shuaiqiang,Liu Tieyan,et al.Listwise collaborative filtering[C] //Proc of the 38th Int ACM SIGIR Conf on Research and Development in Information Retrieval (SIGIR’15).New York:ACM,2015:343- 352
[32]Ma Hao.An experimental study on implicit social recommendation[C] //Proc of the 36th Int ACM SIGIR Conf on Research and Development in Information Retrieval (SIGIR’13).New York:ACM,2013:73- 82
[33]Fazeli S,Loni B,Bellogin A,et al.Implicit vs explicit trust in social matrix factorization[C] //Proc of the 8th ACM Conf on Recommender Systems (RecSys’14).New York:ACM,2014:317- 320
[34]Park C,Kim D,Oh J,et al.Do “also-viewed” products help user rating prediction?[C] //Proc of the 26th Int Conf on World Wide Web (WWW’17).New York:ACM,2017:1113- 1122
① http://www.librec.net