A DEEP LEARNING-BASED RECOMMENDATION SYSTEM FOR TEXTILE PRODUCTS

Abstract

These days, it's more common to use recommendation algorithms to keep customers happy and boost revenue. It is hoped that consumers would be able to make more informed decisions thanks to these solutions. A recommendation system has become an essential aspect of internet buying. Recent emphasis has been focused on fashion and clothes as the subject of several recommendation algorithms. A Convolutional Neural Network (CNN)-based recommendation system has been proposed in this study (CNN). According to the preferences of the CNN's users and designers, distinct patterns have been assigned to different classes in the architecture. Color compatibility is taken into account by the deep learning model when recommending designs for textile items. Our own pattern dataset, which contains 12000 photos, was used to train and evaluate the suggested model. Using pattern datasets, we were able to demonstrate the efficacy of our technique.