摘要
针对保险业对客户信息的分析中缺乏考虑客户信用分析的问题,根据聚类分析算法理论和保险公司客户数据库特点,进一步对K-means聚类算法在大样本环境下初始聚类中心的选取提出有效改进,同时选取一家财产保险公司的客户信用数据,来探讨聚类算法在保险客户信用分析中的应用.
In order to solve the problem that the analysis of policyholders' information lack of considering to analyse policyholders' credit, this paper gives a kind of K-means clustering algorithm that further improved initial center of cluster on environment condition for large sample in terms of cluster analysis theory and trait of policyholders' database, investigates that the algorithm excavates the practice application of policyholders' credit analysis through clients' datum of a property insurance company.
出处
《哈尔滨理工大学学报》
CAS
北大核心
2009年第1期116-119,共4页
Journal of Harbin University of Science and Technology