基于数据挖掘的财险客户风险贡献评级管理

Insurance Customers Segmentation Based on Data Mining-Risk and Contribution Rating Management

良好的客户细分管理能够帮助财险公司更好地管理运营成本与收益,更好地实现公司风险控制和利润最大化的要求。文中采用相关分析进行相关数据的处理,运用K-Means聚类分析、决策树C 5.0算法和改进的Apriori算法3种数据挖掘技术对财险客户从风险和贡献2个角度进行了数据挖掘分类分析,得到具备风险、贡献指向性的双维度客户细分特征变量,并根据这些特征变量,建立了客户风险-贡献分类矩阵,对不同类别的客户提出了不同的客户管理对策建议。

A good customer segmentation management system can help insurance companies administrate the companies＇ costs and benefits better, and it also can help achieve better risk control and profit maximization requirements. This paper, using correlation analysis, K-Means Clustering analysis, Decision Tree C5.0, improved Apriori Algorithm to achieve Data mining classification analysis, and got some variables which were respect of risk and contribution. Finally, the paper established the customers＇ risk-contribution classification matrix, and administrated customers according to different categories.