Bayesian Models for Product Size Recommendations

ABSTRACT

Lack of calibrated product sizing in popular categories such as apparel and shoes leads to customers purchasing incorrect sizes, which in turn results in high return rates due to fit issues. We address the problem of product size recommendations based on customer purchase and return data. We propose a novel approach based on Bayesian logit and probit regression models with ordinal categories Small, Fit, Largeto model size fits as a function of the difference between latent sizes of customers and products. We propose posterior computation based on mean-field variational inference, leveraging the Polya-Gamma augmentation for the logit prior, that results in simple updates, enabling our technique to efficiently handle large datasets. Our Bayesian approach effectively deals with issues arising from noise and sparsity in the data providing robust recommendations. Offline experiments with real-life shoe datasets show that our model outperforms the state-of-the-art in 5 of 6 datasets. and leads to an improvement of 17-26% in AUC over baselines when predicting size fit outcomes.

References

1. Adomavicius, G., and Tuzhilin, A. Context-aware recommender systems. In Recommender systems handbook. Springer, 2011, pp. 217--253.
2. Agarwal, D., and Chen, B.-C. Regression-based latent factor models. In Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (New York, NY, USA, 2009), KDD '09, ACM, pp. 19--28.
3. Aggarwal, C. C. Recommender systems. Springer, 2016.
4. Albert, J. H., and Chib, S. Bayesian analysis of binary and polychotomous response data. Journal of the American statistical Association 88, 422 (1993), 669--679.
5. Bauer, J., and Nanopoulos, A. A framework for matrix factorization based on general distributions. In Proceedings of the 8th ACM Conference on Recommender Systems (New York, NY, USA, 2014), RecSys '14, ACM, pp. 249--256.
6. Bishop, C. M. Pattern Recognition and Machine Learning. Springer, 2006.
7. Charlin, L., Ranganath, R., McInerney, J., and Blei, D. M. Dynamic poisson factorization. In Proceedings of the 9th ACM Conference on Recommender Systems (New York, NY, USA, 2015), RecSys '15, ACM, pp. 155--162.
8. Chen, M.-H., and Dey, D. K. Bayesian analysis for correlated ordinal data models. BIOSTATISTICS-BASEL- 5 (2000), 133--158.
9. Chu, W., and Park, S.-T. Personalized recommendation on dynamic content using predictive bilinear models. In Proceedings of the 18th International Conference on World Wide Web (New York, NY, USA, 2009), WWW '09, ACM, pp. 691--700.
10. Church, K., Smyth, B., Cotter, P., and Bradley, K. Mobile information access: A study of emerging search behavior on the mobile internet. ACM Trans. Web 1, 1 (May 2007).
11. Gopalan, P. K., Charlin, L., and Blei, D. Content-based recommendations with poisson factorization. In Advances in Neural Information Processing Systems (2014), pp. 3176--3184.
12. Greene, W. H., and Hensher, D. A. Modeling ordered choices: A primer. Cambridge University Press, 2010.
13. Grimmer, J. An introduction to bayesian inference via variational approximations. Political Analysis 19, 1 (2010), 32--47.
14. Harvey, M., Carman, M. J., Ruthven, I., and Crestani, F. Bayesian latent variable models for collaborative item rating prediction. In Proceedings of the 20th ACM International Conference on Information and Knowledge Management (New York, NY, USA, 2011), CIKM '11, ACM, pp. 699--708.
15. Joshi, B., Iutzeler, F., and Amini, M.-R. Asynchronous distributed matrix factorization with similar user and item based regularization. In Proceedings of the 10th ACM Conference on Recommender Systems (New York, NY, USA, 2016), RecSys '16, ACM, pp. 75--78.
16. Karatzoglou, A., Amatriain, X., Baltrunas, L., and Oliver, N. Multiverse recommendation: n-dimensional tensor factorization for context-aware collaborative filtering. In Proceedings of the fourth ACM conference on Recommender systems (2010), ACM, pp. 79--86.
17. Kawakatsu, H., and Largey, A. G. EM algorithms for ordered probit models with endogenous regressors. The Econometrics Journal 12, 1 (2009), 164--186.
18. Koren, Y., Bell, R., and Volinsky, C. Matrix factorization techniques for recommender systems. Computer 42, 8 (2009).
19. Liang, D., Altosaar, J., Charlin, L., and Blei, D. M. Factorization meets the item embedding: Regularizing matrix factorization with item co-occurrence. In Proceedings of the 10th ACM conference on recommender systems (2016), ACM, pp. 59--66.
20. McAuley, J., and Leskovec, J. Hidden factors and hidden topics: understanding rating dimensions with review text. In Proceedings of the 7th ACM conference on Recommender systems (2013), ACM, pp. 165--172.
21. McKinley, T. J., Morters, M., Wood, J. L., et al. Bayesian model choice in cumulative link ordinal regression models. Bayesian Analysis 10, 1 (2015), 1--30.
22. Narita, A., Hayashi, K., Tomioka, R., and Kashima, H. Tensor factorization using auxiliary information. Data Mining and Knowledge Discovery 25, 2 (2012), 298--324.
23. Palmisano, C., Tuzhilin, A., and Gorgoglione, M. Using context to improve predictive modeling of customers in personalization applications. IEEE transactions on knowledge and data engineering 20, 11 (2008), 1535--1549.
24. Panniello, U., Tuzhilin, A., Gorgoglione, M., Palmisano, C., and Pedone, A. Experimental comparison of pre-vs. post-filtering approaches in contextaware recommender systems. In Proceedings of the third ACM conference on Recommender systems (2009), ACM, pp. 265--268.
25. Polson, N. G., Scott, J. G., and Windle, J. Bayesian inference for logistic models using pólya--gamma latent variables. Journal of the American statistical Association 108, 504 (2013), 1339--1349.
26. Qin, Z., Rishabh, I., and Carnahan, J. A scalable approach for periodical personalized recommendations. In Proceedings of the 10th ACM Conference on Recommender Systems (New York, NY, USA, 2016), RecSys '16, ACM, pp. 23--26.
27. Rafailidis, D., and Nanopoulos, A. Modeling the dynamics of user preferences in coupled tensor factorization. In Proceedings of the 8th ACM Conference on Recommender Systems (New York, NY, USA, 2014), RecSys '14, ACM, pp. 321--324.
28. Salakhutdinov, R., and Mnih, A. Bayesian probabilistic matrix factorization using markov chain monte carlo. In Proceedings of the 25th International Conference on Machine Learning (New York, NY, USA, 2008), ICML '08, ACM, pp. 880--887.
29. Salakhutdinov, R., and Mnih, A. Probabilistic matrix factorization. In Advances in Neural Information Processing Systems (2008), vol. 20.
30. Sembium, V., Rastogi, R., Saroop, A., and Merugu, S. Recommending product sizes to customers. In Proceedings of the Eleventh ACM Conference on Recommender Systems (New York, NY, USA, 2017), RecSys '17, ACM, pp. 243--250.
31. Vasile, F., Smirnova, E., and Conneau, A. Meta-prod2vec: Product embeddings using side-information for recommendation. In Proceedings of the 10th ACM Conference on Recommender Systems (2016), ACM, pp. 225--232

Index Terms

• Information systems

  • Information retrieval

    • Retrieval tasks and goals

      • Recommender systems

  • Information systems applications

    • Data mining

      • Collaborative filtering

• Mathematics of computing

  • Probability and statistics

    • Probabilistic inference problems

      • Bayesian computation