{"type":"ScholarlyArticle","title":"Learning User Dependencies for Recommendation","authors":[{"name":"Yong Liu","slug":"yong-liu","given":"Yong","family":"Liu","affiliation":"† Institute for Infocomm Research , A"},{"name":"Peilin Zhao","slug":"peilin-zhao","given":"Peilin","family":"Zhao","affiliation":"‡ Artificial Intelligence Department , Ant Financial Services Group , China; * STAR , Singapore"},{"name":"Xin Liu","slug":"xin-liu","given":"Xin","family":"Liu","affiliation":"§ Garena Online , Singapore Big Data Research Center , University of Electronic Science and Technology of China , China"},{"name":"Min Wu","slug":"min-wu","given":"Min","family":"Wu","affiliation":"† Institute for Infocomm Research , A"},{"name":"Lixin Duan","slug":"lixin-duan","given":"Lixin","family":"Duan","affiliation":""},{"name":"Xiao-Li Li","slug":"xiao-li-li","given":"Xiao-Li","family":"Li","affiliation":"† Institute for Infocomm Research , A"}],"abstract":"Social recommender systems exploit users' social relationships to improve recommendation accuracy. Intuitively, a user tends to trust different people regarding with different scenarios. Therefore, one main challenge of social recommendation is to exploit the most appropriate dependencies between users for a given recommendation task. Previous social recommendation methods are usually developed based on pre-defined user dependencies. Thus, they may not be optimal for a specific recommendation task. In this paper, we propose a novel recommendation method, named probabilistic relational matrix factorization (PRMF), which can automatically learn the dependencies between users to improve recommendation accuracy. In PRMF, users' latent features are assumed to follow a matrix variate normal (MVN) distribution. Both positive and negative user dependencies can be modeled by the row precision matrix of the MVN distribution. Moreover, we also propose an alternating optimization algorithm to solve the optimization problem of PRMF. Extensive experiments on four real datasets have been performed to demonstrate the effectiveness of the proposed PRMF model.","doi":"10.24963/ijcai.2017/331","url":"https://ijcai.org/proceedings/2017/331","pdf":"https://ijcai.org/proceedings/2017/0331.pdf","markdown":"https://ijcai.org/proceedings/2017/331.md","bibtex":"https://ijcai.org/proceedings/2017/bibtex/331","volume":{"slug":"2017","title":"Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2017","url":"https://ijcai.org/proceedings/2017/","year":2017,"location":"Melbourne, Australia","dates":"19-25 August 2017","editors":"Carles Sierra","isbn":"978-0-9992411-0-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2017"},"section":"Machine Learning A-R","track":"Main track","number":331,"pages":{"first":2379,"last":2385},"keywords":["Multidisciplinary Topics and Applications: Personalization and User Modeling","Machine Learning: Data Mining"],"videos":[],"references":[{"position":1,"raw":"[Boyd et al., 2011] Stephen Boyd, Neal Parikh, Eric Chu, Borja Peleato, and Jonathan Eckstein. Distributed opti- mization and statistical learning via the alternating direc- tion method of multipliers. Foundations and Trends R in Machine Learning, 3(1):1-122, 2011.","title":"Distributed optimization and statistical learning via the alternating direction method of multipliers","authors":"Stephen Boyd, Neal Parikh, Eric Chu, Borja Peleato, Jonathan Eckstein","year":"2011","venue":"Foundations and Trends R in Machine Learning","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Cai et al., 2011] Tony Cai, Weidong Liu, and Xi Luo. A constrained ℓ 1 minimization approach to sparse precision matrix estimation. Journal of the American Statistical As- sociation, 106(494):594-607, 2011.","title":"A constrained ℓ 1 minimization approach to sparse precision matrix estimation","authors":"Tony Cai, Weidong Liu, Xi Luo","year":"2011","venue":"Journal of the American Statistical Association","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Candes and Tao, 2007] Emmanuel Candes and Terence Tao. The dantzig selector: statistical estimation when p is much larger than n. The Annals of Statistics, pages 2313-2351, 2007.","title":"The dantzig selector: statistical estimation when p is much larger than n","authors":"Emmanuel Candes, Terence Tao","year":"2007","venue":"The Annals of Statistics","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Davis et al., 2007] Jason V Davis, Brian Kulis, Prateek Jain, Suvrit Sra, and Inderjit S Dhillon. Information-theoretic metric learning. In ICML'07, pages 209-216, 2007.","title":"Information-theoretic metric learning","authors":"Jason V Davis, Brian Kulis, Prateek Jain, Suvrit Sra, Inderjit S Dhillon","year":"2007","venue":"ICML'07","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Fang et al., 2014] Hui Fang, Yang Bao, and Jie Zhang. Leveraging decomposed trust in probabilistic matrix fac- torization for effective recommendation. In AAAI'14, pages 30-36, 2014.","title":"Leveraging decomposed trust in probabilistic matrix factorization for effective recommendation","authors":"Hui Fang, Yang Bao, Jie Zhang","year":"2014","venue":"AAAI'14","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Guo et al., 2015] Guibing Guo, Jie Zhang, and Neil Yorke- Smith. Trustsvd: Collaborative filtering with both the ex- plicit and implicit influence of user trust and of item rat- ings. In AAAI'15, pages 123-125, 2015.","title":"Trustsvd: Collaborative filtering with both the explicit and implicit influence of user trust and of item ratings","authors":"Guo","year":"2015","venue":"AAAI'15","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Gupta and Nagar, 1999] Arjun K Gupta and Daya K Nagar. Matrix variate distributions, volume 104. 1999.","title":"Matrix variate distributions","authors":"K Arjun, Gupta, Daya K Nagar","year":"1999","venue":null,"doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Hu et al., 2014] Longke Hu, Aixin Sun, and Yong Liu. Your neighbors affect your ratings: on geographical neighbor- hood influence to rating prediction. In SIGIR'14, pages 345-354. ACM, 2014.","title":"Your neighbors affect your ratings: on geographical neighborhood influence to rating prediction","authors":"Hu","year":"2014","venue":"SIGIR'14","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Hu et al., 2015] Guang-Neng Hu, Xin-Yu Dai, Yunya Song, Shu-Jian Huang, and Jia-Jun Chen. A synthetic approach for recommendation: combining ratings, social relations, and reviews. In IJCAI'15, pages 1756-1762, 2015.","title":"A synthetic approach for recommendation: combining ratings, social relations, and reviews","authors":"Guang-Neng Hu, Xin-Yu Dai, Yunya Song, Shu-Jian Huang, Jia-Jun Chen","year":"2015","venue":"IJCAI'15","doi":null,"url":null,"internal_url":"/proceedings/2015/250"},{"position":10,"raw":"[Jamali and Ester, 2010] Mohsen Jamali and Martin Ester. A matrix factorization technique with trust propagation for recommendation in social networks. In RecSyS'10, pages 135-142. ACM, 2010.","title":"A matrix factorization technique with trust propagation for recommendation in social networks","authors":"Ester ; Jamali, Mohsen Jamali, Martin Ester","year":"2010","venue":"RecSyS'10","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Johnson, 2014] Christopher C Johnson. Logistic matrix fac- torization for implicit feedback data. NIPS'14, 27, 2014.","title":"Logistic matrix factorization for implicit feedback data","authors":"Johnson Christopher","year":"2014","venue":"NIPS'","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Koren et al., 2009] Yehuda Koren, Robert Bell, and Chris Volinsky. Matrix factorization techniques for recom- mender systems. Computer, (8):30-37, 2009.","title":"Matrix factorization techniques for recommender systems","authors":"Koren","year":"2009","venue":"Computer","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Li et al., 2015] Hui Li, Dingming Wu, Wenbin Tang, and Nikos Mamoulis. Overlapping community regularization for rating prediction in social recommender systems. In RecSyS'15, pages 27-34, 2015.","title":"Overlapping community regularization for rating prediction in social recommender systems","authors":"Hui Li, Dingming Wu, Wenbin Tang, Nikos Mamoulis","year":"2015","venue":"RecSyS'15","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Liu and Luo, 2014] Weidong Liu and Xi Luo. High- dimensional sparse precision matrix estimation via sparse column inverse operator. Journal of Multivariate Analysis, 2014.","title":"Highdimensional sparse precision matrix estimation via sparse column inverse operator","authors":"Weidong Liu, Xi Luo","year":"2014","venue":"Journal of Multivariate Analysis","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Liu et al., 2009] Han Liu, John Lafferty, and Larry Wasser- man. The nonparanormal: Semiparametric estimation of high dimensional undirected graphs. Journal of Machine Learning Research, 10(Oct):2295-2328, 2009.","title":"The nonparanormal: Semiparametric estimation of high dimensional undirected graphs","authors":"Han Liu, John Lafferty, Larry Wasserman","year":"2009","venue":"Journal of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Liu et al., 2013] Xin Liu, Yong Liu, Karl Aberer, and Chun- yan Miao. Personalized point-of-interest recommendation by mining users' preference transition. In CIKM'13, pages 733-738, 2013.","title":"Personalized point-of-interest recommendation by mining users' preference transition","authors":"Xin Liu, Yong Liu, Karl Aberer, Chunyan Miao","year":"2013","venue":"CIKM'13","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Liu et al., 2014] Yong Liu, Wei Wei, Aixin Sun, and Chun- yan Miao. Exploiting geographical neighborhood charac- teristics for location recommendation. In CIKM'14, pages 739-748, 2014.","title":"Exploiting geographical neighborhood characteristics for location recommendation","authors":"Yong Liu, Wei Wei, Aixin Sun, Chunyan Miao","year":"2014","venue":"CIKM'14","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Liu et al., 2015] Yong Liu, Peilin Zhao, Aixin Sun, and Chunyan Miao. A boosting algorithm for item recommen- dation with implicit feedback. In IJCAI'15, volume 15, pages 1792-1798, 2015.","title":"A boosting algorithm for item recommendation with implicit feedback","authors":"Yong Liu, Peilin Zhao, Aixin Sun, Chunyan Miao","year":"2015","venue":"IJCAI'15","doi":null,"url":null,"internal_url":"/proceedings/2015/255"},{"position":19,"raw":"[Liu et al., 2016] Yong Liu, Min Wu, Chunyan Miao, Peilin Zhao, and Xiao-Li Li. Neighborhood regularized logistic matrix factorization for drug-target interaction prediction. PLoS Computational Biology, 12(2):e1004760, 2016.","title":"Neighborhood regularized logistic matrix factorization for drug-target interaction prediction","authors":"Yong Liu, Min Wu, Chunyan Miao, Peilin Zhao, Xiao-Li Li","year":"2016","venue":"PLoS Computational Biology","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Ma et al., 2011] Hao Ma, Dengyong Zhou, Chao Liu, Michael R Lyu, and Irwin King. Recommender systems with social regularization. In WSDM'11, pages 287-296, 2011.","title":"Recommender systems with social regularization","authors":"Hao Ma, Dengyong Zhou, Chao Liu, Irwin Michael R Lyu, King","year":"2011","venue":"WSDM'11","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Ma, 2013] Hao Ma. An experimental study on implicit so- cial recommendation. In SIGIR'13, pages 73-82, 2013.","title":"An experimental study on implicit social recommendation","authors":"Hao Ma","year":"2013","venue":"SIGIR'13","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Ma, 2014] Hao Ma. On measuring social friend interest similarities in recommender systems. In SIGIR'14, pages 465-474, 2014.","title":"On measuring social friend interest similarities in recommender systems","authors":"Hao Ma","year":"2014","venue":"SIGIR'14","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Mei et al., 2014] Jian-Ping Mei, Han Yu, Yong Liu, Zhiqi Shen, and Chunyan Miao. A social trust model considering trustees' influence. In PRIMA'14, pages 357-364, 2014.","title":"A social trust model considering trustees' influence","authors":"Jian-Ping Mei, Han Yu, Yong Liu, Zhiqi Shen, Chunyan Miao","year":"2014","venue":"PRIMA'","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Mnih and Salakhutdinov, 2007] Andriy Mnih and Ruslan Salakhutdinov. Probabilistic matrix factorization. In NIPS'07, pages 1257-1264, 2007.","title":"Probabilistic matrix factorization","authors":"Andriy Mnih, Ruslan Salakhutdinov","year":"2007","venue":"NIPS'07","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Su and Khoshgoftaar, 2009] Xiaoyuan Su and Taghi M Khoshgoftaar. A survey of collaborative filtering tech- niques. Advances in Artificial Intelligence, 2009:4, 2009.","title":"A survey of collaborative filtering techniques","authors":"Xiaoyuan Su, M Taghi, Khoshgoftaar","year":"2009","venue":"Advances in Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Tang et al., 2013] Jiliang Tang, Xia Hu, Huiji Gao, and Huan Liu. Exploiting local and global social context for recommendation. In IJCAI'13, pages 2712-2718, 2013.","title":"Exploiting local and global social context for recommendation","authors":"Jiliang Tang, Xia Hu, Huiji Gao, Huan Liu","year":"2013","venue":"IJCAI'13","doi":null,"url":null,"internal_url":"/proceedings/2013/399"},{"position":27,"raw":"[Wang et al., 2013] Huahua Wang, Arindam Banerjee, Cho- Jui Hsieh, Pradeep K Ravikumar, and Inderjit S Dhillon. Large scale distributed sparse precision estimation. In NIPS'13, pages 584-592, 2013.","title":"Large scale distributed sparse precision estimation","authors":"Huahua Wang, Arindam Banerjee, Cho-Jui Hsieh, Pradeep K Ravikumar, Inderjit S Dhillon","year":"2013","venue":"NIPS'13","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Yang et al., 2012] Xiwang Yang, Harald Steck, and Yong Liu. Circle-based recommendation in online social net- works. In KDD'12, pages 1267-1275, 2012.","title":"Circle-based recommendation in online social networks","authors":"Yang","year":"2012","venue":"KDD'12","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Yang et al., 2014] Xiwang Yang, Yang Guo, Yong Liu, and Harald Steck. A survey of collaborative filtering based social recommender systems. Computer Communications, 41:1-10, 2014.","title":"A survey of collaborative filtering based social recommender systems","authors":"Yang","year":"2014","venue":"Computer Communications","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Zhang and Yeung, 2010] Yu Zhang and Dit-Yan Yeung. A convex formulation for learning task relationships in multi-task learning. In UAI'10, pages 733-742, 2010.","title":"A convex formulation for learning task relationships in multi-task learning","authors":"Yeung ; ] Yu Zhang, Dit-Yan Zhang, Yeung","year":"2010","venue":"UAI'10","doi":null,"url":null,"internal_url":null},{"position":31,"raw":"[Zhang et al., 2013] Xingang Zhang, Qijie Gao, Christo- pher S.G. Khoo, and Amos Wu. Categories of friends on social networking sites: An exploratory study. In A- LIEP'13, 2013.","title":"Categories of friends on social networking sites: An exploratory study","authors":"Xingang Zhang, Qijie Gao, S G Christopher, Amos Khoo, Wu","year":"2013","venue":"A-LIEP'13","doi":null,"url":null,"internal_url":null},{"position":32,"raw":"[Zhao et al., 2011] Peilin Zhao, Rong Jin, Tianbao Yang, and Steven C Hoi. Online auc maximization. In ICML'11, pages 233-240, 2011.","title":"Online auc maximization","authors":"Peilin Zhao, Rong Jin, Tianbao Yang, Steven C Hoi","year":"2011","venue":"ICML'11","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}