基于LRFAT模型和改进K-means的汽车忠诚客户细分方法

To realize precise marketing of automobile customers in the era of industrial Internet, it is necessary to cluster and effectively manage customer resources. Aiming at the low number of automobile loyal customers, high potential value and uneven data distribution, an improved customer segmentation LRFAT model based on RFM model was proposed. To improve the accuracy and stability of customer clustering, a method for selecting the initial cluster center of hierarchical K-nearest density peak was proposed with the inspiration of density peaks clustering, and the initial cluster center was selected to optimize K-means. On this basis, the automobile loyal customers were subdivided with improved K-means. Through the automobile sales application of a vehicle manufacturer, the effectiveness of the model and algorithm was verified. At the same time, the detailed analysis was made for different customer groups and the corresponding marketing suggestions were given.