Advancing customer segmentation: A comparative study of hierarchical level weighted method and K-means algorithms

This paper explores the crucial aspect of customer segmentation and its significance in understanding consumer behavior for businesses. Customer segmentation is the process of grouping customers according to their similar characteristics, behavior, and preferences. The study explores the application of a weighted approach and clustering techniques in the customer segmentation process. The supermarket data set used in this research comprises transaction data from a supermarket, encompassing details such as date, time, products purchased, individual product prices, and the total transaction amount. Leveraging this data set, data analysis techniques are employed to gain insights into supermarket customers' behavior. For instance, the relationship between the day of the week and total transaction amounts is studied, factors influencing customer spending patterns are analyzed, and trends in customer behavior are identified. This project focuses on customer segmentation utilizing two distinct approaches: K-means clustering and a weighted method. The first approach involves employing K-means clustering to segment customers based on their characteristics. The second approach utilizes a weighted method, assigning varying importance to specific attributes for enhanced segmentation. The project leverages these two approaches on a data set to gain valuable insights into customer behavior and preferences. By exploring these techniques, businesses can tailor marketing strategies effectively, improve customer engagement, and optimize overall performance for success.