Sentiment based matrix factorization with reliability for recommendation

Abstract Recommender systems aim at predicting users’ preferences based on abundant information, such as user ratings, demographics, and reviews. Although reviews are sparser than ratings, they provide more detailed and reliable information about users’ true preferences. Currently, reviews are often used to improve the explainability of recommender systems. In this paper, we propose the sentiment based matrix factorization with reliability (SBMF+R) algorithm to leverage reviews for prediction. First, we develop a sentiment analysis approach using a new star-based dictionary construction technique to obtain the sentiment score. Second, we design a user reliability measure that combines user consistency and the feedback on reviews. Third, we incorporate the ratings, reviews, and feedback into a probabilistic matrix factorization framework for prediction. Experiments on eight Amazon datasets demonstrated that SBMF+R is more accurate than state-of-the-art algorithms.

[1]  Lei Zhang,et al.  Sentiment Analysis and Opinion Mining , 2017, Encyclopedia of Machine Learning and Data Mining.

[2]  Jie Zhang,et al.  TopicMF: Simultaneously Exploiting Ratings and Reviews for Recommendation , 2014, AAAI.

[3]  Yue Lu,et al.  Automatic construction of a context-aware sentiment lexicon: an optimization approach , 2011, WWW.

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

[5]  Andrea Esuli,et al.  SentiWordNet 3.0: An Enhanced Lexical Resource for Sentiment Analysis and Opinion Mining , 2010, LREC.

[6]  Alexander J. Smola,et al.  Jointly modeling aspects, ratings and sentiments for movie recommendation (JMARS) , 2014, KDD.

[7]  Alok N. Choudhary,et al.  Voice of the Customers: Mining Online Customer Reviews for Product Feature-based Ranking , 2010, WOSN.

[8]  Barry Smyth,et al.  Combining similarity and sentiment in opinion mining for product recommendation , 2015, Journal of Intelligent Information Systems.

[9]  Clement T. Yu,et al.  The effect of negation on sentiment analysis and retrieval effectiveness , 2009, CIKM.

[10]  Huan Liu,et al.  Context-aware review helpfulness rating prediction , 2013, RecSys.

[11]  Janyce Wiebe,et al.  Recognizing Contextual Polarity in Phrase-Level Sentiment Analysis , 2005, HLT.

[12]  Elena Lloret,et al.  Experiments on Summary-based Opinion Classification , 2010, HLT-NAACL 2010.

[13]  Woo Jun Kim,et al.  Propagation Loss of Line-Defect Photonic Crystal Slab Waveguides , 2006, IEEE Journal of Selected Topics in Quantum Electronics.

[14]  Lorenzo Bruzzone,et al.  Definition of Effective Training Sets for Supervised Classification of Remote Sensing Images by a Novel Cost-Sensitive Active Learning Method , 2014, IEEE Transactions on Geoscience and Remote Sensing.

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

[16]  Pearl Pu,et al.  Prediction of Helpful Reviews Using Emotions Extraction , 2014, AAAI.

[17]  Yehuda Koren,et al.  Matrix Factorization Techniques for Recommender Systems , 2009, Computer.

[18]  Guokun Lai,et al.  Explicit factor models for explainable recommendation based on phrase-level sentiment analysis , 2014, SIGIR.

[19]  Li Chen,et al.  News impact on stock price return via sentiment analysis , 2014, Knowl. Based Syst..

[20]  David Schuff,et al.  Is It the Review or the Reviewer? a Multi-Method Approach to Determine the Antecedents of Online Review Helpfulness , 2011, 2011 44th Hawaii International Conference on System Sciences.

[21]  Björn W. Schuller,et al.  New Avenues in Opinion Mining and Sentiment Analysis , 2013, IEEE Intelligent Systems.

[22]  Heyan Huang,et al.  A Method of Polarity Computation of Chinese Sentiment Words Based on Gaussian Distribution , 2014, CICLing.

[23]  Hong Yu,et al.  Tag recommendation method in folksonomy based on user tagging status , 2017, Journal of Intelligent Information Systems.

[24]  William Hart-Davidson,et al.  A method for measuring helpfulness in online peer review , 2010, SIGDOC '10.

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

[26]  Heyan Huang,et al.  Labeled Phrase Latent Dirichlet Allocation , 2016, WISE.

[27]  Shenghuo Zhu,et al.  SumView: A Web-based engine for summarizing product reviews and customer opinions , 2013, Expert Syst. Appl..

[28]  Kerstin Denecke,et al.  Using SentiWordNet for multilingual sentiment analysis , 2008, 2008 IEEE 24th International Conference on Data Engineering Workshop.

[29]  Zhu Wang,et al.  A sentiment-enhanced personalized location recommendation system , 2013, HT.

[30]  Soo-Min Kim,et al.  Automatically Assessing Review Helpfulness , 2006, EMNLP.

[31]  H. Sebastian Seung,et al.  Algorithms for Non-negative Matrix Factorization , 2000, NIPS.

[32]  Xiao-Jun Zeng,et al.  ISTS: Implicit social trust and sentiment based approach to recommender systems , 2015, Expert Syst. Appl..

[33]  Iryna Gurevych,et al.  Beyond the stars: exploiting free-text user reviews to improve the accuracy of movie recommendations , 2009, TSA@CIKM.

[34]  Andrei Popescu-Belis,et al.  Sentiment analysis of user comments for one-class collaborative filtering over ted talks , 2013, SIGIR.

[35]  Paul A. Pavlou,et al.  Can online reviews reveal a product's true quality?: empirical findings and analytical modeling of Online word-of-mouth communication , 2006, EC '06.

[36]  Li Chen,et al.  Sentiment-enhanced explanation of product recommendations , 2014, WWW.

[37]  Mohan S. Kankanhalli,et al.  Aspect-Aware Latent Factor Model: Rating Prediction with Ratings and Reviews , 2018, WWW.

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

[39]  Huan Liu,et al.  Exploiting social relations for sentiment analysis in microblogging , 2013, WSDM.

[40]  Siamak Faridani.  Using canonical correlation analysis for generalized sentiment analysis, product recommendation and search , 2011, RecSys '11.

[41]  Michael J. Pazzani,et al.  Content-Based Recommendation Systems , 2007, The Adaptive Web.

[42]  Amélie Marian,et al.  Beyond the Stars: Improving Rating Predictions using Review Text Content , 2009, WebDB.

[43]  Li Chen,et al.  Recommender systems based on user reviews: the state of the art , 2015, User Modeling and User-Adapted Interaction.

[44]  Li Chen,et al.  Comparison of feature-level learning methods for mining online consumer reviews , 2012, Expert Syst. Appl..

[45]  Jianwu Dang,et al.  Twitter summarization with social-temporal context , 2016, World Wide Web.

[46]  Xiaohui Yu,et al.  Sentiment analysis of sentences with modalities , 2013, UnstructureNLP@CIKM.

[47]  Bamshad Mobasher,et al.  Towards Trustworthy Recommender Systems : An Analysis of Attack Models and Algorithm Robustness , 2007 .

[48]  Mitsuru Ishizuka,et al.  SentiFul: A Lexicon for Sentiment Analysis , 2011, IEEE Transactions on Affective Computing.

[49]  Maite Taboada,et al.  Lexicon-Based Methods for Sentiment Analysis , 2011, CL.

[50]  Joydeep Ghosh,et al.  Review quality aware collaborative filtering , 2012, RecSys '12.