A Survey on Fashion Recommendation Systems

Abstract

This study focuses on a recommendation system for a fashion company that sells online and offline products the type of products are similar in both ways, either online or offline and the quality of products is based on seasons. Product evolution is based on customer choice. Customers mostly buy products if they like and want to repurchase and the second is to try something new which is not on their list the demand for products is changing according to customer behavior. This research proposed a precise review of fashion recommendation systems for the collaborative filtering (CF) approach which identifies and filtered out the selling of products in higher quantity on both online and offline scales. It also filters the demand of products with respect to change in time and uses Singular Value Decomposition (SVD) for item-based factorization to find the most hit and visited products for future consideration and to improve its accuracy. This review focuses on the appropriate screening techniques and the various models that can be used in the development of referral systems.