【摘要】 针对数据密集的物流企业客户细分问题,提出了一种基于稀疏模糊聚类的物流客户细分方法。在数据预处理阶段,结合物流客户价值特点对传统RFM模型进行改进,构建了物流客户价值指标体系,提高了物流聚类细分数据集合的准确度和全面性。在聚类细分阶段,引入稀疏聚类方法改进FCM算法进行特征选择并提高聚类方法性能,最终得到物流客户细分结果。实例测试及与其他算法比较表明,该方法有更好的细分效果和稳健性,可以有效识别出不同价值的物流客户群体,为企业决策分析提供依据。更多还原

【Abstract】 For the customer segmentation problem of logistics enterprises with intensive data, a new method of logistics customer segmentation based on sparse fuzzy clustering algorithm is proposed. An index system of logistics customer value based on characteristics of logistics customers and traditional RFM model is constructed to improve the accuracy and comprehensiveness of data sets. In the phase of clustering, the FCM algorithm based on sparse clustering method is developed to achieve the selection of characteristic variables and improve the performance of the clustering algorithm. The results of the logistics customer segmentation are obtained. The experimental results show that the method has better segmentation effect and can identify the different value customer groups effectively, which provides a basis for the decision analysis of the enterprise.更多还原