基于用户评论的深度情感分析和多视图协同融合的混合推荐方法
|
摘要
目前,大多数推荐技术使用用户评分来推断用户偏好.当有充足的评分信息时,协同过滤技术表现良好.然而,评分数据普遍存在着稀疏性,或者难以让用户将其偏好表示为对物品的评分等级,故有效性受到限制.基于内容的推荐方法依据物品的内容来寻找与目标用户喜欢的物品内容相似的物品.在目标用户没有充足的历史数据的情况下,该方法仍然不充分,其推荐效果也很有限.当前,融合多视图的兴趣偏好信息构建混合推荐系统是个性化推荐研究发展的趋势.混合推荐系统通过融合用户物品的交互评分、隐式反馈和辅助信息进行个性化推荐,故本文提出了一种新颖的基于用户评论的深度情感分析和多视图协同融合的混合推荐方法.针对用户评论、物品内容描述等短文本的情感及语义难以分析,单一推荐视图易导致对用户画像建模粗放等问题,本文利用词向量对用户评论的短文本进行分布式表征,并结合长短期记忆网络实现从上下文语义层面对用户评论的情感进行分析.同时,本文提出基于观点预过滤和基于用户评分嵌入的情感融合方法,设计了一种嵌入的网络结构对用户评论进行深层语义分析和情感计算,以解决用户评分与真实兴趣偏好存在较大偏差、评分等级分布极度不均衡等问题.此外,本文利用分布式的段落向量表征对物品内容描述的短文本进行相似度计算,并设计了候选物品相似性的计算方法及度量K个最近邻物品的方法,解决了推荐系统中物品的内容信息不易挖掘和利用的问题.最后,本文提出了一种基于协同训练的融合用户评分、情感倾向和物品内容信息的混合推荐算法,实现对稀疏的用户评分矩阵的循环填充和修正,进而实现基于评分预测的TopN推荐.该方法解决了混合推荐系统中不同兴趣偏好的多推荐视图难以融合的问题,同时在一定程度上解决了推荐系统建模中缺乏足够的有标签数据问题.本文在亚马逊数据集上进行实验,与多种经典的和当前先进的推荐算法进行性能对比,采用平方误差、命中率和标准化折扣累积增益进行性能评价.实验结果表明,本文提出的算法在挖掘用户情感上效果显著;在10个推荐数据集上,系统的评分预测和TopN推荐指标皆有不同程度的显著改进.
Currently,most recommender techniques use user ratings to infer user preferences.Collaborative filtering techniques perform well when there is sufficient rating information.However,their effectiveness is limited because of the rating sparsity problem,or the difficulty in letting users express their preferences as scalar ratings on items.Content-based recommender methods rely instead on the content representations of items to locate items that have similar content to items the target user liked.However,these methods are still inadequate and its recommendation effect is limited,especially when the target user has little historical data.At present,it is a recent development trend to do personalized recommendation through fusing multi-view of interest preferences to build the hybrid recommendation model,which usually makes personalized recommendation with user-item interaction ratings,implicit feedback and auxiliary information in hybrid recommendation system.In this paper,a novel hybrid recommendation algorithm is proposed that based on deep sentiment analysis of user reviews and multi-view collaborative fusion.For these problems that it is difficult to analyze user reviews' sentiment and items content'semantics,and a single view of the recommended model lead to user profile is extensive,we use Word2 vec to characterize the short texts of user reviews and combine long short-term memory networks to realize the sentiment analysis of the user review on the context semantic level.At the same time,a sentiment fusion method based on opinion pre-filtering and user rating embedding is proposed,and an embedded network structure is designed for deep semantic analysis and sentiment calculation of user's review.The proposed method will solve the problem that there is a great deviation between the user's rating and real interest preference,and also solve the extreme imbalance problem of the user rating distribution.In addition,we use the distributed vector representation of paragraph to characterize the short text of the item's text description,so as to realize the similarity calculation of the item's content.We design a method to measure the similarity of candidate items and calculate K nearest neighbor items,which solves the problem that the item's content information is not easy to mine and use in recommendation system.Finally,a fusion method of recommendation view based on collaborative training is proposed,which integrates user ratings,sentiment preferences and item's content information.It can fill and modify the sparse user ratings matrix,and then realize recommendation based on ratings prediction.It solves the problem that multi-recommendation views with different interests and preferences are difficult to fuse in hybrid recommendation system,and solves the problem of lack of sufficient labeled data for modeling in a certain degree.We conduct the experiments on Amazon product dataset,and compare our algorithm with a variety of classic and state-of-the-art recommendation algorithms.Specially,the results are evaluated in Mean Squared Error,Hit Radio,and Normalized Discounted Cumulative Gain.The experiment result shows that the algorithm proposed in this paper has a significant effect in mining user's sentiment.On the ten recommended datasets,our algorithm has also a significant improvement in the accuracy of the score prediction and TopNperformance of the recommendation system in different degrees.
Currently,most recommender techniques use user ratings to infer user preferences.Collaborative filtering techniques perform well when there is sufficient rating information.However,their effectiveness is limited because of the rating sparsity problem,or the difficulty in letting users express their preferences as scalar ratings on items.Content-based recommender methods rely instead on the content representations of items to locate items that have similar content to items the target user liked.However,these methods are still inadequate and its recommendation effect is limited,especially when the target user has little historical data.At present,it is a recent development trend to do personalized recommendation through fusing multi-view of interest preferences to build the hybrid recommendation model,which usually makes personalized recommendation with user-item interaction ratings,implicit feedback and auxiliary information in hybrid recommendation system.In this paper,a novel hybrid recommendation algorithm is proposed that based on deep sentiment analysis of user reviews and multi-view collaborative fusion.For these problems that it is difficult to analyze user reviews' sentiment and items content'semantics,and a single view of the recommended model lead to user profile is extensive,we use Word2 vec to characterize the short texts of user reviews and combine long short-term memory networks to realize the sentiment analysis of the user review on the context semantic level.At the same time,a sentiment fusion method based on opinion pre-filtering and user rating embedding is proposed,and an embedded network structure is designed for deep semantic analysis and sentiment calculation of user's review.The proposed method will solve the problem that there is a great deviation between the user's rating and real interest preference,and also solve the extreme imbalance problem of the user rating distribution.In addition,we use the distributed vector representation of paragraph to characterize the short text of the item's text description,so as to realize the similarity calculation of the item's content.We design a method to measure the similarity of candidate items and calculate K nearest neighbor items,which solves the problem that the item's content information is not easy to mine and use in recommendation system.Finally,a fusion method of recommendation view based on collaborative training is proposed,which integrates user ratings,sentiment preferences and item's content information.It can fill and modify the sparse user ratings matrix,and then realize recommendation based on ratings prediction.It solves the problem that multi-recommendation views with different interests and preferences are difficult to fuse in hybrid recommendation system,and solves the problem of lack of sufficient labeled data for modeling in a certain degree.We conduct the experiments on Amazon product dataset,and compare our algorithm with a variety of classic and state-of-the-art recommendation algorithms.Specially,the results are evaluated in Mean Squared Error,Hit Radio,and Normalized Discounted Cumulative Gain.The experiment result shows that the algorithm proposed in this paper has a significant effect in mining user's sentiment.On the ten recommended datasets,our algorithm has also a significant improvement in the accuracy of the score prediction and TopNperformance of the recommendation system in different degrees.
引文
[1]Cremonesi P,Koren Y,Turrin R.Performance of recommender algorithms on top-N recommendation tasks//Proceedings of the 4th ACM Conference on Recommender Systems.Barcelona,Spain,2010:39-46
[2]Wang H,Wang N,Yeung D Y.Collaborative deep learning for recommender systems//Proceedings of the 21st ACMSIGKDD International Conference on Knowledge Discovery and Data Mining.Sydney,Australia,2015:1235-1244
[3]Wei J,He J,Chen K,et al.Collaborative filtering and deep learning based recommendation system for cold start items.Expert Systems with Applications,2017,69:29-39
[4]Bobadilla J,Ortega F,Hernando A,et al.Recommender systems survey.Knowledge-Based Systems,2013,46:109-132
[5]Hu Zhong-Kai,Zheng Xiao-Lin,Wu Ya-Feng,et al.Product recommendation algorithm based on user’s reviews mining.Journal of Zhejiang University(Engineering Science),2013,47(8):1475-1485(in Chinese)(扈中凯,郑小林,吴亚峰等.基于用户评论挖掘的产品推荐算法.浙江大学学报(工学版),2013,47(8):1475-1485)
[6]Wang Zhi-Sheng,Li Qi,Wang Jing,et al.Real-time personalized recommendation based on implicit user feedback data stream.Chinese Journal of Computers,2016,39(1):52-64(in Chinese)(王智圣,李琪,汪静等.基于隐式用户反馈数据流的实时个性化推荐.计算机学报,2016,39(1):52-64)
[7]Zhang F,Yuan N J,Lian D,et al.Collaborative knowledge base embedding for recommender systems//Proceedings of the22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.San Francisco,USA,2016:353-362
[8]Huang Zhen-Hua,Zhang Jia-Wen,Tian Chun-Qi,et al.Survey on learning-to-rank based recommendation algorithm.Journal of Software,2016,27(3):691-713(in Chinese)(黄震华,张佳雯,田春岐等.基于排序学习的推荐算法研究综述.软件学报,2016,27(3):691-713)
[9]Karatzoglou A,Baltrunas L,Shi Y.Learning to rank for recommender systems//Proceedings of the 7th ACM Conference on Recommender Systems.Hong Kong,China,2013:493-494
[10]Zhang Y,Zhang H,Zhang M,et al.Do users rate or review?:Boost phrase-level sentiment labeling with review-level sentiment classification//Proceedings of the 37th International ACMSIGIR Conference on Research&Development in Information Retrieval.Gold Coast,Australia,2014:1027-1030
[11]Chen Long,Guan Zi-Yu,He Jin-Hong,et al.A survey on sentiment classification.Journal of Computer Research and Development,2017,54(6):1150-1170(in Chinese)(陈龙,管子玉,何金红等.情感分类研究进展.计算机研究与发展,2017,54(6):1150-1170)
[12]Zhang W,Wang J.A collective Bayesian Poisson factorization model for cold-start local event recommendation//Proceedings of the 21st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.Sydney,Australia,2015:1455-1464
[13]Huang Li-Wei,Jiang Bi-Tao,Lv Shou-Ye,et al.Survey on deep learning based recommender systems.Chinese Journal of Computers,2018,41(7):1619-1647(in Chinese)(黄立威,江碧涛,吕守业等.基于深度学习的推荐系统研究综述.计算机学报,2018,41(7):1619-1647)
[14]Li Lin,Liu Jin-Hang,Meng Xiang-Fu,et al.Recommendation models by exploiting rating matrix and review text.Chinese Journal of Computers,2018,41(7):1559-1573(in Chinese)(李琳,刘锦行,孟祥福等.融合评分矩阵与评论文本的商品推荐模型.计算机学报,2018,41(7):1559-1573)
[15]Xie Xin-Qiang,Yang Xiao-Chun,Wang Bin,et al.Multifeature fused software developer recommendation.Journal of Software,2018,29(8):2306-2321(in Chinese)(谢新强,杨晓春,王斌等.一种多特征融合的软件开发者推荐.软件学报,2018,29(8):2306-2321)
[16]Pan Yi-Teng,He Fa-Zhi,Yu Hai-Ping.Social recommendation algorithm using implicit similarity in trust.Chinese Journal of Computers,2018,41(1):65-81(in Chinese)(潘一腾,何发智,于海平.一种基于信任关系隐含相似度的社会化推荐算法.计算机学报,2018,41(1):65-81)
[17]Kong Xin-Xin,Su Ben-Chang,Wang Hong-Zhi,et al.Research on the modeling and related algorithms of label-weight rating based recommendation system.Chinese Journal of Computers,2017,40(6):1440-1452(in Chinese)(孔欣欣,苏本昌,王宏志等.基于标签权重评分的推荐模型及算法研究.计算机学报,2017,40(6):1440-1452)
[18]Tian Chao,Qin Zuo-Yan,Zhu Qing,et al.SuperRank:An intelligent recommendation system based on review analysis.Journal of Computer Research and Development,2010,47(1):494-498(in Chinese)(田超,覃左言,朱青等.SuperRank:基于评论分析的智能推荐系统.计算机研究与发展,2010,47(1):494-498)
[19]Shmueli E,Kagian A,Koren Y,et al.Care to comment?:Recommendations for commenting on news stories//Proceedings of the 21st International Conference on World Wide Web.Lyon,France,2012:429-438
[20]Zhang Y,Tan Y,Zhang M,et al.Catch the black sheep:Unified framework for shilling attack detection based on fraudulent action propagation//Proceedings of the 24th International Joint Conference on Artificial Intelligence.Buenos Aires,Argentina,2015:2408-2414
[21]Wu Y,DuBois C,Zheng A X,et al.Collaborative denoising auto-encoders for top-N recommender systems//Proceedings of the 9th ACM International Conference on Web Search and Data Mining.San Francisco,USA,2016:153-162
[22]Chen L,Chen G,Wang F.Recommender systems based on user reviews:the state of the art.User Modeling and UserAdapted Interaction,2015,25(2):99-154
[23]Zhang W,Yuan Q,Han J,et al.Collaborative multi-level embedding learning from reviews for rating prediction//Proceedings of the 25th International Joint Conference on Artificial Intelligence.New York,USA,2016:2986-2992
[24]Zheng L,Noroozi V,Yu P S.Joint deep modeling of users and items using reviews for recommendation//Proceedings of the 10th ACM International Conference on Web Search and Data Mining.Cambridge,United Kingdom,2017:425-434
[25]Chen C,Zhang M,Liu Y,et al.Neural attentional rating regression with review-level explanations//Proceedings of the2018 World Wide Web Conference on World Wide Web.Lyon,France,2018:1583-1592
[26]Han X,Shi C,Wang S,et al.Aspect-level deep collaborative filtering via heterogeneous information networks//Proceedings of the 27th International Joint Conference on Artificial Intelligence.Stockholm,Sweden,2018:3393-3399
[27]Liu Zhi-Yuan,Sun Mao-Song,Lin Yan-Kai,et al.Knowledge representation learning:A review.Journal of Computer Research and Development,2016,53(2):1-16(in Chinese)(刘知远,孙茂松,林衍凯等.知识表示学习研究进展.计算机研究与发展,2016,53(2):1-16)
[28]Wang P,Xu J,Xu B,et al.Semantic clustering and convolutional neural network for short text categorization//Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics.Beijing,China,2015:352-357
[29]Wang Zhong-Yuan,Chen Jian-Peng,Wang Hai-Xun,Wen Ji-Rong.Short text understanding:A survey.Journal of Computer Research and Development,2016,53(2):262-269(in Chinese)(王仲远,陈健鹏,王海勋,文继荣.短文本理解研究.计算机研究与发展,2016,53(2):262-269)
[30]Ma W,Zhang M,Wang C,et al.Your tweets reveal what you like:Introducing cross-media content information into multi-domain recommendation//Proceedings of the 27th International Joint Conference on Artificial Intelligence.Stockholm,Sweden,2018:3484-3490
[31]Lian J,Zhang F,Xie X,et al.Towards better representation learning for personalized news recommendation:A multichannel deep fusion approach//Proceedings of the 27th International Joint Conference on Artificial Intelligence.Stockholm,Sweden,2018:3805-3811
[32]Nguyen T T,Hui P M,Harper F M,et al.Exploring the filter bubble:The effect of using recommender systems on content diversity//Proceedings of the 23rd International Conference on World Wide Web.Seoul,Korea,2014:677-686
[33]Wu Z,Wu J,Cao J,et al.HySAD:A semi-supervised hybrid shilling attack detector for trustworthy product recommendation//Proceedings of the 18th ACM SIGKDDInternational Conference on Knowledge Discovery and Data Mining.Beijing,China,2012:985-993
[34]Zhang M,Tang J,Zhang X,et al.Addressing cold start in recommender systems:A semi-supervised co-training algorithm//Proceedings of the 37th International ACM SIGIR Conference on Research&Development in Information Retrieval.Gold Coast,Australia,2014:73-82
[35]Ding J,Yu G,He X,et al.Improving implicit recommender systems with view data//Proceedings of the 27th International Joint Conference on Artificial Intelligence.Stockholm,Sweden,2018:3343-3349
[36]Wang X,He X,Feng F,et al.TEM:Tree-enhanced embedding model for explainable recommendation//Proceedings of the2018 World Wide Web Conference on World Wide Web.Lyon,France,2018:1543-1552
[37]Pero,Horvth T.Opinion-driven matrix factorization for rating prediction//Proceedings of the 21st International Conference on User Modeling,Adaptation,and Personalization.Rome,Italy,2013:1-13
[38]Mikolov T,Sutskever I,Chen K,et al.Distributed representations of words and phrases and their compositionality//Proceedings of the Advances in Neural Information Processing Systems.Lake Tahoe,USA,2013:3111-3119
[39]Le Q,Mikolov T.Distributed representations of sentences and documents//Proceedings of the International Conference on Machine Learning.Beijing,China,2014:1188-1196
[40]McAuley J,Leskovec J.Hidden factors and hidden topics:Understanding rating dimensions with review text//Proceedings of the 7th ACM Conference on Recommender Systems.Hong Kong,China,2013:165-172
[41]Qiu L,Gao S,Cheng W,et al.Aspect-based latent factor model by integrating ratings and reviews for recommender system.Knowledge-Based Systems,2016,110:233-243
[42]Sarwar B,Karypis G,Konstan J,et al.Item-based collaborative filtering recommendation algorithms//Proceedings of the 10th International Conference on World Wide Web.Hong Kong,China,2001:285-295
[43]Koren Y,Bell R,Volinsky C.Matrix factorization techniques for recommender systems.Computer,2009,42(8):42-49
[44]Koren Y.Factorization meets the neighborhood:A multifaceted collaborative filtering model//Proceedings of the 14th ACMSIGKDD International Conference on Knowledge Discovery and Data Mining.Las Vegas,USA,2008:426-434
[45]Xue H J,Dai X,Zhang J,et al.Deep matrix factorization models for recommender systems//Proceedings of the 26th International Joint Conference on Artificial Intelligence.Melbourne,Australia,2017:3203-3209
[46]He X,Liao L,Zhang H,et al.Neural collaborative filtering//Proceedings of the 26th International Conference on World Wide Web.Perth,Australia,2017:173-182
[47]Gantner Z,Rendle S,Freudenthaler C,et al.MyMediaLite:A free recommender system library//Proceedings of the 5th ACM Conference on Recommender Systems.Chicago,USA,2011:305-308
[2]Wang H,Wang N,Yeung D Y.Collaborative deep learning for recommender systems//Proceedings of the 21st ACMSIGKDD International Conference on Knowledge Discovery and Data Mining.Sydney,Australia,2015:1235-1244
[3]Wei J,He J,Chen K,et al.Collaborative filtering and deep learning based recommendation system for cold start items.Expert Systems with Applications,2017,69:29-39
[4]Bobadilla J,Ortega F,Hernando A,et al.Recommender systems survey.Knowledge-Based Systems,2013,46:109-132
[5]Hu Zhong-Kai,Zheng Xiao-Lin,Wu Ya-Feng,et al.Product recommendation algorithm based on user’s reviews mining.Journal of Zhejiang University(Engineering Science),2013,47(8):1475-1485(in Chinese)(扈中凯,郑小林,吴亚峰等.基于用户评论挖掘的产品推荐算法.浙江大学学报(工学版),2013,47(8):1475-1485)
[6]Wang Zhi-Sheng,Li Qi,Wang Jing,et al.Real-time personalized recommendation based on implicit user feedback data stream.Chinese Journal of Computers,2016,39(1):52-64(in Chinese)(王智圣,李琪,汪静等.基于隐式用户反馈数据流的实时个性化推荐.计算机学报,2016,39(1):52-64)
[7]Zhang F,Yuan N J,Lian D,et al.Collaborative knowledge base embedding for recommender systems//Proceedings of the22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.San Francisco,USA,2016:353-362
[8]Huang Zhen-Hua,Zhang Jia-Wen,Tian Chun-Qi,et al.Survey on learning-to-rank based recommendation algorithm.Journal of Software,2016,27(3):691-713(in Chinese)(黄震华,张佳雯,田春岐等.基于排序学习的推荐算法研究综述.软件学报,2016,27(3):691-713)
[9]Karatzoglou A,Baltrunas L,Shi Y.Learning to rank for recommender systems//Proceedings of the 7th ACM Conference on Recommender Systems.Hong Kong,China,2013:493-494
[10]Zhang Y,Zhang H,Zhang M,et al.Do users rate or review?:Boost phrase-level sentiment labeling with review-level sentiment classification//Proceedings of the 37th International ACMSIGIR Conference on Research&Development in Information Retrieval.Gold Coast,Australia,2014:1027-1030
[11]Chen Long,Guan Zi-Yu,He Jin-Hong,et al.A survey on sentiment classification.Journal of Computer Research and Development,2017,54(6):1150-1170(in Chinese)(陈龙,管子玉,何金红等.情感分类研究进展.计算机研究与发展,2017,54(6):1150-1170)
[12]Zhang W,Wang J.A collective Bayesian Poisson factorization model for cold-start local event recommendation//Proceedings of the 21st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.Sydney,Australia,2015:1455-1464
[13]Huang Li-Wei,Jiang Bi-Tao,Lv Shou-Ye,et al.Survey on deep learning based recommender systems.Chinese Journal of Computers,2018,41(7):1619-1647(in Chinese)(黄立威,江碧涛,吕守业等.基于深度学习的推荐系统研究综述.计算机学报,2018,41(7):1619-1647)
[14]Li Lin,Liu Jin-Hang,Meng Xiang-Fu,et al.Recommendation models by exploiting rating matrix and review text.Chinese Journal of Computers,2018,41(7):1559-1573(in Chinese)(李琳,刘锦行,孟祥福等.融合评分矩阵与评论文本的商品推荐模型.计算机学报,2018,41(7):1559-1573)
[15]Xie Xin-Qiang,Yang Xiao-Chun,Wang Bin,et al.Multifeature fused software developer recommendation.Journal of Software,2018,29(8):2306-2321(in Chinese)(谢新强,杨晓春,王斌等.一种多特征融合的软件开发者推荐.软件学报,2018,29(8):2306-2321)
[16]Pan Yi-Teng,He Fa-Zhi,Yu Hai-Ping.Social recommendation algorithm using implicit similarity in trust.Chinese Journal of Computers,2018,41(1):65-81(in Chinese)(潘一腾,何发智,于海平.一种基于信任关系隐含相似度的社会化推荐算法.计算机学报,2018,41(1):65-81)
[17]Kong Xin-Xin,Su Ben-Chang,Wang Hong-Zhi,et al.Research on the modeling and related algorithms of label-weight rating based recommendation system.Chinese Journal of Computers,2017,40(6):1440-1452(in Chinese)(孔欣欣,苏本昌,王宏志等.基于标签权重评分的推荐模型及算法研究.计算机学报,2017,40(6):1440-1452)
[18]Tian Chao,Qin Zuo-Yan,Zhu Qing,et al.SuperRank:An intelligent recommendation system based on review analysis.Journal of Computer Research and Development,2010,47(1):494-498(in Chinese)(田超,覃左言,朱青等.SuperRank:基于评论分析的智能推荐系统.计算机研究与发展,2010,47(1):494-498)
[19]Shmueli E,Kagian A,Koren Y,et al.Care to comment?:Recommendations for commenting on news stories//Proceedings of the 21st International Conference on World Wide Web.Lyon,France,2012:429-438
[20]Zhang Y,Tan Y,Zhang M,et al.Catch the black sheep:Unified framework for shilling attack detection based on fraudulent action propagation//Proceedings of the 24th International Joint Conference on Artificial Intelligence.Buenos Aires,Argentina,2015:2408-2414
[21]Wu Y,DuBois C,Zheng A X,et al.Collaborative denoising auto-encoders for top-N recommender systems//Proceedings of the 9th ACM International Conference on Web Search and Data Mining.San Francisco,USA,2016:153-162
[22]Chen L,Chen G,Wang F.Recommender systems based on user reviews:the state of the art.User Modeling and UserAdapted Interaction,2015,25(2):99-154
[23]Zhang W,Yuan Q,Han J,et al.Collaborative multi-level embedding learning from reviews for rating prediction//Proceedings of the 25th International Joint Conference on Artificial Intelligence.New York,USA,2016:2986-2992
[24]Zheng L,Noroozi V,Yu P S.Joint deep modeling of users and items using reviews for recommendation//Proceedings of the 10th ACM International Conference on Web Search and Data Mining.Cambridge,United Kingdom,2017:425-434
[25]Chen C,Zhang M,Liu Y,et al.Neural attentional rating regression with review-level explanations//Proceedings of the2018 World Wide Web Conference on World Wide Web.Lyon,France,2018:1583-1592
[26]Han X,Shi C,Wang S,et al.Aspect-level deep collaborative filtering via heterogeneous information networks//Proceedings of the 27th International Joint Conference on Artificial Intelligence.Stockholm,Sweden,2018:3393-3399
[27]Liu Zhi-Yuan,Sun Mao-Song,Lin Yan-Kai,et al.Knowledge representation learning:A review.Journal of Computer Research and Development,2016,53(2):1-16(in Chinese)(刘知远,孙茂松,林衍凯等.知识表示学习研究进展.计算机研究与发展,2016,53(2):1-16)
[28]Wang P,Xu J,Xu B,et al.Semantic clustering and convolutional neural network for short text categorization//Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics.Beijing,China,2015:352-357
[29]Wang Zhong-Yuan,Chen Jian-Peng,Wang Hai-Xun,Wen Ji-Rong.Short text understanding:A survey.Journal of Computer Research and Development,2016,53(2):262-269(in Chinese)(王仲远,陈健鹏,王海勋,文继荣.短文本理解研究.计算机研究与发展,2016,53(2):262-269)
[30]Ma W,Zhang M,Wang C,et al.Your tweets reveal what you like:Introducing cross-media content information into multi-domain recommendation//Proceedings of the 27th International Joint Conference on Artificial Intelligence.Stockholm,Sweden,2018:3484-3490
[31]Lian J,Zhang F,Xie X,et al.Towards better representation learning for personalized news recommendation:A multichannel deep fusion approach//Proceedings of the 27th International Joint Conference on Artificial Intelligence.Stockholm,Sweden,2018:3805-3811
[32]Nguyen T T,Hui P M,Harper F M,et al.Exploring the filter bubble:The effect of using recommender systems on content diversity//Proceedings of the 23rd International Conference on World Wide Web.Seoul,Korea,2014:677-686
[33]Wu Z,Wu J,Cao J,et al.HySAD:A semi-supervised hybrid shilling attack detector for trustworthy product recommendation//Proceedings of the 18th ACM SIGKDDInternational Conference on Knowledge Discovery and Data Mining.Beijing,China,2012:985-993
[34]Zhang M,Tang J,Zhang X,et al.Addressing cold start in recommender systems:A semi-supervised co-training algorithm//Proceedings of the 37th International ACM SIGIR Conference on Research&Development in Information Retrieval.Gold Coast,Australia,2014:73-82
[35]Ding J,Yu G,He X,et al.Improving implicit recommender systems with view data//Proceedings of the 27th International Joint Conference on Artificial Intelligence.Stockholm,Sweden,2018:3343-3349
[36]Wang X,He X,Feng F,et al.TEM:Tree-enhanced embedding model for explainable recommendation//Proceedings of the2018 World Wide Web Conference on World Wide Web.Lyon,France,2018:1543-1552
[37]Pero,Horvth T.Opinion-driven matrix factorization for rating prediction//Proceedings of the 21st International Conference on User Modeling,Adaptation,and Personalization.Rome,Italy,2013:1-13
[38]Mikolov T,Sutskever I,Chen K,et al.Distributed representations of words and phrases and their compositionality//Proceedings of the Advances in Neural Information Processing Systems.Lake Tahoe,USA,2013:3111-3119
[39]Le Q,Mikolov T.Distributed representations of sentences and documents//Proceedings of the International Conference on Machine Learning.Beijing,China,2014:1188-1196
[40]McAuley J,Leskovec J.Hidden factors and hidden topics:Understanding rating dimensions with review text//Proceedings of the 7th ACM Conference on Recommender Systems.Hong Kong,China,2013:165-172
[41]Qiu L,Gao S,Cheng W,et al.Aspect-based latent factor model by integrating ratings and reviews for recommender system.Knowledge-Based Systems,2016,110:233-243
[42]Sarwar B,Karypis G,Konstan J,et al.Item-based collaborative filtering recommendation algorithms//Proceedings of the 10th International Conference on World Wide Web.Hong Kong,China,2001:285-295
[43]Koren Y,Bell R,Volinsky C.Matrix factorization techniques for recommender systems.Computer,2009,42(8):42-49
[44]Koren Y.Factorization meets the neighborhood:A multifaceted collaborative filtering model//Proceedings of the 14th ACMSIGKDD International Conference on Knowledge Discovery and Data Mining.Las Vegas,USA,2008:426-434
[45]Xue H J,Dai X,Zhang J,et al.Deep matrix factorization models for recommender systems//Proceedings of the 26th International Joint Conference on Artificial Intelligence.Melbourne,Australia,2017:3203-3209
[46]He X,Liao L,Zhang H,et al.Neural collaborative filtering//Proceedings of the 26th International Conference on World Wide Web.Perth,Australia,2017:173-182
[47]Gantner Z,Rendle S,Freudenthaler C,et al.MyMediaLite:A free recommender system library//Proceedings of the 5th ACM Conference on Recommender Systems.Chicago,USA,2011:305-308