Spectrum-enhanced Pairwise Learning to Rank

To enhance the performance of the recommender system, side information is extensively explored with various features (e.g., visual features and textual features). However, there are some demerits of side information: (1) the extra data is not always available in all recommendation tasks; (2) it is only for items, there is seldom high-level feature describing users. To address these gaps, we introduce the spectral features extracted from two hypergraph structures of the purchase records. Spectral features describe the similarity of users/items in the graph space, which is critical for recommendation. We leverage spectral features to model the users' preference and items' properties by incorporating them into a Matrix Factorization (MF) model. In addition to modeling, we also use spectral features to optimize. Bayesian Personalized Ranking (BPR) is extensively leveraged to optimize models in implicit feedback data. However, in BPR, all missing values are regarded as negative samples equally while many of them are indeed unseen positive ones. We enrich the positive samples by calculating the similarity among users/items by the spectral features. The key ideas are: (1) similar users shall have similar preference on the same item; (2) a user shall have similar perception on similar items. Extensive experiments on two real-world datasets demonstrate the usefulness of the spectral features and the effectiveness of our spectrum-enhanced pairwise optimization. Our models outperform several state-of-the-art models significantly.

[1]  Dongwon Lee,et al.  “Told you i didn't like it”: Exploiting uninteresting items for effective collaborative filtering , 2016, 2016 IEEE 32nd International Conference on Data Engineering (ICDE).

[2]  J. G. Lewis,et al.  A Shifted Block Lanczos Algorithm for Solving Sparse Symmetric Generalized Eigenproblems , 1994, SIAM J. Matrix Anal. Appl..

[3]  Tommy W. S. Chow,et al.  A Hypergraph Model for Incorporating Social Interactions in Collaborative Filtering , 2017, DMCIT '17.

[4]  Xiangnan He,et al.  Attentive Collaborative Filtering: Multimedia Recommendation with Item- and Component-Level Attention , 2017, SIGIR.

[5]  Daniel P. W. Ellis,et al.  Content-Aware Collaborative Music Recommendation Using Pre-trained Neural Networks , 2015, ISMIR.

[6]  Julian J. McAuley,et al.  VBPR: Visual Bayesian Personalized Ranking from Implicit Feedback , 2015, AAAI.

[7]  Lars Schmidt-Thieme,et al.  BPR: Bayesian Personalized Ranking from Implicit Feedback , 2009, UAI.

[8]  Xu Chen,et al.  Aesthetic-based Clothing Recommendation , 2018 .

[9]  John Riedl,et al.  Item-based collaborative filtering recommendation algorithms , 2001, WWW '01.

[10]  Tie-Yan Liu,et al.  Learning to rank: from pairwise approach to listwise approach , 2007, ICML '07.

[11]  Congfu Xu,et al.  Adaptive Bayesian personalized ranking for heterogeneous implicit feedbacks , 2015, Knowl. Based Syst..

[12]  Bernhard Schölkopf,et al.  Learning with Hypergraphs: Clustering, Classification, and Embedding , 2006, NIPS.

[13]  Tat-Seng Chua,et al.  Neural Factorization Machines for Sparse Predictive Analytics , 2017, SIGIR.

[14]  Stefano Battiston,et al.  A model of a trust-based recommendation system on a social network , 2006, Autonomous Agents and Multi-Agent Systems.

[15]  Yehuda Koren,et al.  The BellKor Solution to the Netflix Grand Prize , 2009 .

[16]  Douglas B. Terry,et al.  Using collaborative filtering to weave an information tapestry , 1992, CACM.

[17]  Hanqing Lu,et al.  Item group based pairwise preference learning for personalized ranking , 2014, SIGIR.

[18]  Max Welling,et al.  Graph Convolutional Matrix Completion , 2017, ArXiv.

[19]  Qiang Yang,et al.  Matrix Factorization+ for Movie Recommendation , 2016, IJCAI.

[20]  Michael I. Jordan,et al.  On Spectral Clustering: Analysis and an algorithm , 2001, NIPS.

[21]  Lionel Martin,et al.  Fast Eigenspace Approximation using Random Signals , 2016, ArXiv.

[22]  Tat-Seng Chua,et al.  Fast Matrix Factorization for Online Recommendation with Implicit Feedback , 2016, SIGIR.

[23]  Jure Leskovec,et al.  Hidden factors and hidden topics: understanding rating dimensions with review text , 2013, RecSys.

[24]  Jun Wang,et al.  Optimizing top-n collaborative filtering via dynamic negative item sampling , 2013, SIGIR.

[25]  M. Sharir,et al.  A strong-connectivity algorithm and its applications in data flow analysis. , 2018 .

[26]  Xiaoyu Du,et al.  Outer Product-based Neural Collaborative Filtering , 2018, IJCAI.

[27]  Andrew P. Bradley,et al.  The use of the area under the ROC curve in the evaluation of machine learning algorithms , 1997, Pattern Recognit..

[28]  Nadia Magnenat-Thalmann,et al.  Time-aware point-of-interest recommendation , 2013, SIGIR.

[29]  Ruslan Salakhutdinov,et al.  Probabilistic Matrix Factorization , 2007, NIPS.

[30]  Tat-Seng Chua,et al.  Neural Collaborative Filtering , 2017, WWW.

[31]  Li Chen,et al.  Proceedings of the Twenty-Third International Joint Conference on Artificial Intelligence GBPR: Group Preference Based Bayesian Personalized Ranking for One-Class Collaborative Filtering , 2022 .

[32]  Tao Chen,et al.  Context-aware Image Tweet Modelling and Recommendation , 2016, ACM Multimedia.

[33]  Lei Zheng,et al.  Spectral collaborative filtering , 2018, RecSys.

[34]  Jiawei Han,et al.  Non-negative Matrix Factorization on Manifold , 2008, 2008 Eighth IEEE International Conference on Data Mining.

[35]  Pascal Frossard,et al.  The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains , 2012, IEEE Signal Processing Magazine.

[36]  Xiangnan He,et al.  NAIS: Neural Attentive Item Similarity Model for Recommendation , 2018, IEEE Transactions on Knowledge and Data Engineering.

[37]  Depeng Jin,et al.  An Improved Sampler for Bayesian Personalized Ranking by Leveraging View Data , 2018, WWW.

[38]  Tong Zhao,et al.  Leveraging Social Connections to Improve Personalized Ranking for Collaborative Filtering , 2014, CIKM.

[39]  Dongjin Song,et al.  Recommending Positive Links in Signed Social Networks by Optimizing a Generalized AUC , 2015, AAAI.

[40]  Steffen Rendle,et al.  Improving pairwise learning for item recommendation from implicit feedback , 2014, WSDM.

[41]  Anton van den Hengel,et al.  Image-Based Recommendations on Styles and Substitutes , 2015, SIGIR.

[42]  Martin Ester,et al.  A matrix factorization technique with trust propagation for recommendation in social networks , 2010, RecSys '10.

[43]  Pradeep Ravikumar,et al.  Collaborative Filtering with Graph Information: Consistency and Scalable Methods , 2015, NIPS.

[44]  Yiqun Liu,et al.  Learning on Partial-Order Hypergraphs , 2018, WWW.

[45]  Xing Zhang,et al.  CPLR: Collaborative pairwise learning to rank for personalized recommendation , 2018, Knowl. Based Syst..

[46]  Huanbo Luan,et al.  Discrete Collaborative Filtering , 2016, SIGIR.

[47]  Yi Xiong,et al.  List-wise probabilistic matrix factorization for recommendation , 2014, Inf. Sci..

[48]  James Bennett,et al.  The Netflix Prize , 2007 .

[49]  Yongfeng Zhang,et al.  Personalized Key Frame Recommendation , 2017, SIGIR.

[50]  Mao Ye,et al.  Exploiting geographical influence for collaborative point-of-interest recommendation , 2011, SIGIR.

[51]  Steffen Rendle,et al.  Factorization Machines , 2010, 2010 IEEE International Conference on Data Mining.