A Product Recommendation System for a gift shop web application

Abstract

E-commerce websites have been noticed over the recent years due to their benefits for sellers and buyers, and improving them has become a need to keep the business in the competition. The main objective of this thesis is to build an e-commerce website for a gift shop that provides best shopping experience to customers, including recommending the products that match their needs based on the customers purchase history. This study derived from the profound importance of recommendation systems in improving the customer experience on one hand, and raising the sales and upsell of merchants and businesses on the other hand, in addition to personalizing the suggestions for each customer as it positively affects customers loyalty, retention and satisfaction. The thesis focused on applying some machine learning tools in addition to the selling statistics to suggest products. As a result, an e-commerce web application was implemented (in PHP, MySQL, and Pyhton with ML library "Tensorflow") to provide shopping functionalities and recommend products to users using a model that can predict similar items to the customer’s interests.