ABSTRACT
Reading online reviews before a purchase is a customary action nowadays. Nevertheless, the increasing volume of reviews works as a barrier to their effectiveness so that many approaches try to predict reviews' quality, which is not standardized to all users due to different backgrounds and preferences. Thus, recommending reviews in a personalized fashion is probably more accurate. Here, we analyze methods for recommending reviews that have not been compared against each other yet. Our experiments consider parameter tuning and comparison through statistical tests. Such study allows to understand the state-of-the-art and to evidence potential improvement directions. Our results show that assessing under a ranking perspective, model simplicity and observed features are important traits for this problem, being Support Vector Regression the best solution.
- G. Adomavicius and A. Tuzhilin. Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions. IEEE Trans. on Knowl. and Data Eng., 17(6):734--749, 2005. Google ScholarDigital Library
- D. Agarwal and B.-C. Chen. Regression-based latent factor models. In Procs. of KDD, Paris, France, 2009. Google ScholarDigital Library
- M. Anderson. Local consumer review survey 2014, 2014. https://www.brightlocal.com/2014/07/01/local-consumer-review-survey-2014.Google Scholar
- M. L. Anderson and J. R. Magruder. Learning from the crowd: Regression discontinuity estimates of the effects of an online review database. Economic Journal, 122(563):957--989, 2012.Google ScholarCross Ref
- S. Balakrishnan and S. Chopra. Collaborative ranking. In Procs. of WSDM, Seattle, WA, USA, 2012. Google ScholarDigital Library
- K. Balog and H. Ramampiaro. Cumulative citation recommendation: Classification vs. ranking. In Procs. of SIGIR, Dublin, Ireland, 2013. Google ScholarDigital Library
- J. G. Booth and J. P. Hobert. Maximizing generalized linear mixed model likelihoods with an automated monte carlo em algorithm. JRRS: B, 61(1):265--285, 1999.Google Scholar
- C. J. Burges. From RankNet to LambdaRank to LambdaMART: An overview. Technical Report MSR-TR-2010--82, Microsoft Research, 2010.Google Scholar
- S. D. Canuto et al. A comparative study of learning-to-rank techniques for tag recommendation. JIDM, 4(3):453--468, 2013.Google Scholar
- O. Chapelle and Y. Chang. Yahoo! learning to rank challenge overview. In Procs. of the Yahoo! Learning to Rank Challenge - ICML, Bellevue, WA, USA, 2011.Google Scholar
- O. Chapelle and S. S. Keerthi. Efficient algorithms for ranking with SVMs. IR, 13(3):201--215, 2010. Google ScholarDigital Library
- O. Chapelle, D. Metlzer, Y. Zhang, and P. Grinspan. Expected reciprocal rank for graded relevance. In Procs. of CIKM, Hong Kong, China, 2009. Google ScholarDigital Library
- P. Cremonesi, Y. Koren, and R. Turrin. Performance of recommender algorithms on top-n recommendation tasks. In Procs. of RecSys, Barcelona, Spain, 2010. Google ScholarDigital Library
- J. H. Friedman. Greedy function approximation: A gradient boosting machine. The Annals of Statistics, 29(5):pp. 1189--1232, 2001.Google ScholarCross Ref
- S. Holm. A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics, 6(2):65--70, 1979.Google Scholar
- R. Jain. The art of computer systems performance analysis - techniques for experimental design, measurement, simulation, and modeling. Wiley, 1991.Google Scholar
- T. Joachims. Training linear SVMs in linear time. In Procs. of KDD, Philadelphia, PA, USA, 2006. Google ScholarDigital Library
- Y. Koren. Factorization meets the neighborhood: A multifaceted collaborative filtering model. In Procs. of KDD, Las Vegas, NV, USA, 2008. Google ScholarDigital Library
- Y. Koren, R. Bell, and C. Volinsky. Matrix factorization techniques for recommender systems. Commun. ACM, 42(8):30--37, 2009. Google ScholarDigital Library
- S. Lee and J. Y. Choeh. Predicting the helpfulness of online reviews using multilayer perceptron neural networks. Expert Syst. Appl., 41(6):3041--3046, 2014. Google ScholarDigital Library
- H. Li. A short introduction to learning to rank. IEICE Transactions, 94-D(10):1854--1862, 2011.Google Scholar
- Y. Liu, X. Huang, A. An, and X. Yu. Modeling and predicting the helpfulness of online reviews. In Procs. of ICDM, Miami, FL, USA, 2008. Google ScholarDigital Library
- W. Lu, S. Chen, K. Li, and L. V. S. Lakshmanan. Show me the money: Dynamic recommendations for revenue maximization. PVLDB, 7(14):1785--1796, 2014. Google ScholarDigital Library
- Y. Lu et al. Exploiting social context for review quality prediction. In Procs. of WWW, Raleigh, NC, USA, 2010. Google ScholarDigital Library
- C. Lucchese et al. Quickscorer: A fast algorithm to rank documents with additive ensembles of regression trees. In Procs. of SIGIR, Santiago, Chile, 2015. Google ScholarDigital Library
- A. Machanavajjhala, A. Korolova, and A. D. Sarma. Personalized social recommendations: Accurate or private. PVLDB, 4(7):440--450, Apr. 2011. Google ScholarDigital Library
- L. Martin and P. Pu. Prediction of helpful reviews using emotions extraction. In Procs. of AAAI, Québec City, Canada, 2014. Google ScholarDigital Library
- G. A. Miller. WordNet: A lexical database for english. Commun. ACM, 38(11):39--41, 1995. Google ScholarDigital Library
- S. Moghaddam, M. Jamali, and M. Ester. ETF: Extended Tensor Factorization model for personalizing prediction of review helpfulness. In Procs. of the WSDM, Seattle, WA, USA, 2012. Google ScholarDigital Library
- K. P. Murphy. Machine Learning: A Probabilistic Perspective. The MIT Press, 2012. Google ScholarDigital Library
- F. e. a. Pedregosa. Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12(Oct):2825--2830, 2011. Google ScholarDigital Library
- F. Ricci et al. Recommender Systems Handbook. Springer-Verlag, 1st edition, 2010. Google ScholarDigital Library
- Y. Shi, M. Larson, and A. Hanjalic. Collaborative filtering beyond the user-item matrix: A survey of the state of the art and future challenges. ACM Comput. Surv., 47(1):3:1--3:45, 2014. Google ScholarDigital Library
- A. J. Smola and B. Schölkopf. A tutorial on support vector regression. Statistics and Computing, 14(3):199--222, 2004. Google ScholarDigital Library
- J. Tang et al. Context-aware review helpfulness rating prediction. In Procs. of RecSys, Hong Kong, China, 2013. Google ScholarDigital Library
- V. N. Vapnik. The Nature of Statistical Learning Theory. Springer-Verlag, 1995. Google ScholarDigital Library
- M. Vasconcelos, J. M. Almeida, and M. A. Gonçalves. Predicting the popularity of micro-reviews: A foursquare case study. Information Sciences, 325:355--374, 2015. Google ScholarDigital Library
- H. Yin et al. Challenging the long tail recommendation. PVLDB, 5(9):896--907, 2012. Google ScholarDigital Library
- T. J. Ypma. Historical development of the Newton-Raphson method. SIAM Rev., 37(4):531--551, Dec. 1995. Google ScholarDigital Library
- K. Zhang, R. Narayanan, and A. Choudhary. Voice of the customers: Mining online customer reviews for product feature-based ranking. In Procs. of the WOSN, 2010. Google ScholarDigital Library
- R. Zhang and T. Tran. An information gain-based approach for recommending useful product reviews. Knowl. Inf. Syst., 26(3):419--434, 2011.Google ScholarDigital Library
Index Terms
Assessing Review Recommendation Techniques under a Ranking Perspective
Recommendations
Pairwise preference regression for cold-start recommendation
RecSys '09: Proceedings of the third ACM conference on Recommender systemsRecommender systems are widely used in online e-commerce applications to improve user engagement and then to increase revenue. A key challenge for recommender systems is providing high quality recommendation to users in ``cold-start" situations. We ...
Unifying rating-oriented and ranking-oriented collaborative filtering for improved recommendation
We propose a novel unified recommendation model, URM, which combines a rating-oriented collaborative filtering (CF) approach, i.e., probabilistic matrix factorization (PMF), and a ranking-oriented CF approach, i.e., list-wise learning-to-rank with ...
Multi-Task Learning with Personalized Transformer for Review Recommendation
Web Information Systems Engineering – WISE 2021AbstractDrastic increase in item/product review volume has caused serious information overload. Traditionally, product reviews are exhibited in chronological or popularity order without personalization. The review recommendation model provides users with ...
Comments