System for styling apparels that match with occasion wearing via E-commerce channel using machine learning

Abstract

This work studies the factors effecting the selection of apparels according to fashion styling via an e-commerce platform in order to build an automatic system for suggesting proper apparel concurred with the customers’ preferences. The factors obtained from interviewing experts in the fashion area found that cloth items, apparel colors, skin colors, and body shape are the main factors effecting the selection of fashion styling via an e-commerce platform. These discovered factors were used to partition a set of experimental fashion styling apparel and also to construct a set of suggesting rules based on different values of each factor. The experimental results based on various values of each factor concluded that the accuracy of suggested apparel according to the factors and suggesting rules concurred with the customers’ preference more than 80%.