An Approach towards Deployable Hybrid Product Recommendation Systems for E-Commerce

Abstract

As India is moving fast towards digital economy, E-commerce industry has been on rise. Many platforms provide their customers with a shopping experience better than actual physical stores. Several E-commerce websites use different methods to improve the customer engagement and revenue. One such technique is the use of personalized recommendation systems, which uses customer’s data like interests, purchase history, ratings to suggest new products, which they may like. Recom-mendation systems are used by E-commerce websites to suggest new products to their users. The products can be suggested based on the top merchants on the website, based on the interests of the user or based the past purchase pattern of the cus-tomer. Recommender systems are machine learning based systems that help users discover new products. Due to the recent pandemic situation of 2020 and 2021, many of the local retail stores have been trying to shift their business to online plat-forms such as dedicated websites or social media. The proposed methodology based on Machine Learning aims to enable local online retail business owners to enhance their customer engagement and revenue by providing users with personalized recommendations using past data using methods such as Collaborative Filtering, Popularity-based and Content-Based Filtering.