Enabling Customer Segmentation Based on Classical Partitioning Methods Upon Statistical Evaluation for E-commerce Business Applications

In the present age, the prevailing spirit is centered around innovation, driving individuals to compete with one another. For businesses, the primary objective is not centered around acquiring new customers, but rather on how to maximize sales to their existing customer base. In the competitive field of e-commerce, it is essential to satisfy the customer’s needs and identify the potential customer. This research project delves into advanced customer segmentation techniques using clustering algorithms, specifically K-means, K-medoids, Agglomerative Nesting (AGNES), and Density-Based Spatial Clustering of Applications with Noise (DBSCAN). These clustering methods, coupled with meticulous hyperparameter tuning, enable us to uncover distinct customer groups within e-commerce businesses. By doing so, we gain invaluable insights into the unique characteristics, preferences, and behaviors of each customer segment. Through a comprehensive analysis of these segments, we aim to facilitate the development of highly targeted and personalized marketing strategies.