K-Means Clustering Approach for Intelligent Customer Segmentation Using Customer Purchase Behavior Data

Abstract

E-commerce system has become more popular and implemented in almost all business areas. E-commerce system is a platform for marketing and promoting the products to customer through online. Customer segmentation is known as a process of dividing the customers into groups which shares similar characteristics. The purpose of customer segmentation is to determine how to deal with customers in each category in order to increase the profit of each customer to the business. Segmenting the customers assist business to identify their profitable customer to satisfy their needs by optimizing the services and products. Therefore, customer segmentation helps E-commerce system to promote the right product to the right customer with the intention to increase profits. There are few types of customer segmentation factors which are demographic psychographic, behavioral, and geographic. In this study, customer behavioral factor has been focused. Therefore users will be analyzed using clustering algorithm in determining the purchase behavior of E-commerce system. The aim of clustering is to optimize the experimental similarity within the cluster and to maximize the dissimilarity in between clusters. In this study there are relationship between three clusters: event type, products, and categories. In this research, the proposed approach analyzed the groups that share similar criteria to help vendors to identify and focus on the high profitable segment to the least profitable segment. This type of analysis can play important role in improving the business. Grouping their customer according to their similar behavioral factor to sustain their customer for long-term and increase their business profit. It also enables high exposure of the e-offer to gain attention of potential customers. In order to process the collected data and segment the customers, an learning algorithm is used which is known as K-Means clustering. K-Means clustering is implemented to solve the clustering problems.