与k均值混合的支持向量机的个人信用评估  

Personal credit scoring based on hybrid support vector regression with k-means method

作　　者：高尚[1] 刘夫成[1] 

机构地区：[1]江苏科技大学计算机科学与工程学院,江苏镇江212003

出　　处：《中南大学学报（自然科学版）》2013年第S2期169-173,共5页Journal of Central South University:Science and Technology

基　　金：人工智能四川省重点实验室开放基金资助项目(2009RY001);"青蓝工程"资助项目(2010)

摘　　要：针对当前传统个人信用评价中的问题,提出一种将k-mean聚类方法和支持向量机回归模型结合起来的个人信用评估的新方法。该方法先将信用数据大致聚为k类,统计每一类的信用度,然后利用支持向量机进行回归。研究结果表明:与简单聚类方法和支持向量机分类方法比较,该方法有效地提高了整个模型的训练精度和测试精度。To solve the problems in traditional personal credit scoring system,an individual credit assessment model based on hybrid support vector regression with k-means method was proposed.The personal credit data were clustered to k kinds by k-means,then the personal credit scoring of every kind was counted,and the credit data were regressed by support vector regression(SVR).The results show that compared with simple clustering method and support vector classification method,the proposed approach improves training precision and test precision of the whole model.

关 键 词：个人信用评估 K均值 支持向量机 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] O212.1[自动化与计算机技术—控制科学与工程]