针对客户市场细分问题进行了研究。依据粗糙集理论,以信息表中条件属性与决策属性的一致性原理为基础,以超立方体、扫描向量为数据计算对象,进行数据离散化和连续型属性约简,实现了数据预处理;在此基础上,以集合差异度指标为约束条件,运用集合特征向量加法法则最终实现客户市场细分。在实验中,连续属性离散化和冗余属性约简有效地减少了计算数据,便于客户市场细分的实现,提高了客户市场细分的效果。研究结果表明该客户市场细分算法是有效可行的。 Research on customer segmentation. According to the theory of rough set theory, based on the consistency principle of conditional attributes and decision attributes in information table, hypercube and scan vector are used as data computing objects, data discretization and continuous attribute reduction are realized, and data preprocessing is realized. In Based on this, taking the set diversity index as the constraint, the customer e-market segmentation is finally achieved by using the set eigenvector addition method. In the experiment, continuous attribute discretization and redundancy attribute reduction effectively reduce the calculation data, facilitate the realization of customer market segmentation and improve the effect of customer segmentation. The results show that the customer segmentation algorithm is effective and feasible.