Customer Segmentation For Scalable Recommender Systems in Online Retailing

Abstract

Recommender Systems are an important tool used by companies and businesses in order to gain profit via customer satisfaction. It’s becoming increasingly popular among online retailers and often considered as a crucial strategy to attract customers. Retailers who understand the value of recommendations are looking to improve the quality of recommendations in any possible way. Nowadays, with the increment of users, products and online business interactions, massive volumes of data are being stored and processed by recommender systems. This led to large scale online retailers making their recommender systems to be scalable and efficient with Big data technologies. Customer Segmentation is an important approach taken by organizations in order to find their most profitable customers. Retailers are keen on identifying their valuable and nonvaluable customers and offer customized services to them. This project focuses on incorporating customer segmentation with recommendation algorithms in order to improve the quality of recommendations. The solution provided in this project is intended to be a part of a recommender system and is compatible with the architecture and design of modern-day scalable recommendation engines. Low cost and complexity, as well as scalability, are some of the key characteristics of the solution while the outcome is much more relevant and diverse recommendations.