| 基于SVM的房贷信用评估的应用研究 |
| |
| 引用本文： | 王波,郝艳友,刘勇奎,刘爽.基于SVM的房贷信用评估的应用研究[J].计算机工程与设计,2008,29(19). |
| |
| 作者姓名： | 王波 郝艳友 刘勇奎 刘爽 |
| |
| 作者单位： | 1. 大连民族学院,计算机科学与工程学院,辽宁,大连,116605
2. 大连理工大学,计算机科学与工程系,辽宁,大连,116001 |
| |
| 摘 要： | 信贷风险是金融机构风险主要来源.支持向量机(SVM)在解决两类问题上是一种较好的分类方法,其学习模型有较强的稳定性.对SVM在房贷信用评估应用中的问题进行了研究和解决,如核函数选取,参数选取,样本非均衡问题等.实验得出在实际应用中径向基模型较好,采用Grid-search方法调整参数,能达到更好的推广能力和预测结果,用分别惩罚支持向量机能有效解决样本非均衡问题.试验结果也证明了基于SVM的房贷信用评估方法优于原有的打分方法.
|
| 关 键 词： | 信用评估 支持向量机 网格搜索 参数选取 非均衡样本 |
Application research of housing loan credit evaluation based on SVM |
| |
| WANG Bo,HAO Yan-you,LIU Yong-kui,LIU Shuang.Application research of housing loan credit evaluation based on SVM[J].Computer Engineering and Design,2008,29(19). |
| |
| Authors: | WANG Bo HAO Yan-you LIU Yong-kui LIU Shuang |
| |
| Affiliation: | WANG Bo1,HAO Yan-you2,LIU Yong-kui1,LIU Shuang1(1.College of Computer Science , Engineering,Dalian Nationalities University,Dalian 116605,China,2.Department of Computer Science , Engineering,Dalian University of Technology,Dalian 116001,China) |
| |
| Abstract: | Credit risk is the primary source of risk to financial institutions.Support vector machine(SVM) is a good classifier to solve binary classification problem and the learning results possess stronger robustness.Resolving the problem of the application of SVM in housing loan credit evaluation,such as the choice of kernel function and parameters,the problem of unbalance data.The experiment show RBF model is suitable better for practical application,grid-search method adjusts these penalty parameters to achieve ... |
| |
| Keywords: | credit rating support vector machine grid-search parameters selection unbalanced data |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
