Design and Analysis of Apparel Recommendation System

This paper presented a deterministic apparel recommendation system designed to overcome the critical limitations of existing probabilistic approaches. Bycollecting body measurements at registration, the system delivers fit- guaranteedproductrecommendationstoallusers—including newuserswithnopurchasehistory—fullyresolvingthecold- start problem. The unified cross-retailer size normalization framework ensures recommendations remain consistent and physically accurate regardless of the source retailer. The proposed system represents a fundamental shift from predicting what a user might like to deterministicallyidentifying whatwillfit.Evaluationconfirmssignificantimprovementsin fit accuracy and cross-retailer consistency, validating the practicality of the approach for real-world e-commerce deployment.

1.  Goldberg, D., Nichols, D., Oki, B. M., Terry, D.: Using collaborative filtering to weave an information tapestry. Communications of the ACM, 35(12), 61–70 (1992) 2.   Koren, Y., Bell, R., Volinsky, C.: Matrix factorization techniques for recommender systems. IEEE Computer, 42(8), 30–37 (2009) 3.  Ashdown, S. P.: Sizing in Clothing: Developing Effective Sizing Systems for Ready-to-Wear Clothing. Woodhead Publishing, Cambridge (2007) 4.  He, R., McAuley, J.: Ups and Downs: Modeling the visual evolution of fashion trends with one-class collaborative filtering. In: Proceedings of WWW, pp. 507–517 (2016) 5.  Schein, A. I., Popescul, A., Ungar, L. H., Pennock, D. M.: Methods and metrics for cold-start recommendations. In: Proceedings of ACM SIGIR, pp. 253–260 (2002)