CUSTOMER SEGMENTATION OF E-COMMERCE INDUSTRY BASED ON RSA MODEL AND IMPROVED K-MEANS ALGORITHM

Aimed at the characteristics of the emerging online shopping customers, such as large number of customers, strong customer mobility and large consumption data, the RSA model combined with the improved K-means clustering algorithm is proposed to achieve customer segmentation. The entropy method was used to calculate the weight of each index of the RSA model, and the customer value was calculated by integrating each attribute. Combined with K-nearest neighbor algorithm and density peak algorithm, a K-means initial cluster center selection method based on K-nearest neighbor and density peak clustering was proposed, and the traditional K-means algorithm was optimized to achieve customer segmentation. The selected standard data set and the real data of a retail company’sonline transaction were verified by experiments. It is proved that the RSA model and the improved K-means algorithm have more excellent performance.