An Online Recommendation System Using Deep Learning for Textile Products

Abstract

Recommendation systems are frequently preferred in recent years ensuring customer satisfaction and accelerating sales. Thanks to these systems, it is aimed to accelerate the decision-making process of customers. Recommendation systems have become a necessary part, especially in online shopping. Most of the recommendation systems used in many different areas have been attracting attention, focusing on fashion, and clothing recently. In this paper, a deep learning-based online recommendation system has been proposed with a Convolutional Neural Network (CNN). Classes of different patterns in the CNN architecture have been determined according to users' and designers' pattern preferences. The deep learning model recommends patterns considering color compatibility for textile products. The proposed model has been trained and tested using our own pattern dataset including 12000 images. Experiments on pattern datasets show the effectiveness of our proposed approach.