A Machine Learning Approach to Identify Potential Customer Based on Purchase Behavior

Abstract

To lift the revenue boundary and stay ahead of the competitors it is important to understand customer’s purchase behavior. Different business industries proposed different policies to explore the potentiality of a customer based on statistical analysis. In this paper, we rather propose a machine learning approach to identify potential customers for a retail superstore. The paper proposed an engineered approach to classify potential customer, based on previously recorded purchase behavior. Using this classification as ground truth, we then apply machine learning algorithms to find a pattern to predict potential customers with an accuracy of 99.4%.