E-commerce Customer Segmentation and Precision Marketing Strategy Combined with Cluster Analysis

E-commerce platforms usually have a large amount of data from different channels and systems, which have problems of different quality and insufficient consistency. In addition, segmentation based on customer data relies too much on some simple rules and models, resulting in a rough division of customer groups and ignoring the multidimensional characteristics and individual needs of customers. Therefore, this study combines cluster analysis to conduct relevant research on e-commerce customer segmentation and precision marketing strategies, aiming to solve the problems of incomplete data, oversimplification of customer segmentation, and lack of personalized recommendations in the research of e-commerce customer segmentation and precision marketing strategies. The study uses data fusion algorithms to process data from different platforms, and uses density-based spatial clustering of applications with noise (DBSCAN), K-means, and hierarchical clustering (HC) to refine customer groups from multiple angles, providing personalized recommendations and precision marketing for different user groups. Experiments show that the method combining K-means, hierarchical clustering, and DBSCAN has an ARI (Adjusted Rand Index) of 0.91, a noise processing capability of 0.93, and a calculation time of 8.7 s, which is better than a single clustering method.