支持向量机在企业信用评价中的应用研究  

Study of Support Vector Machine in Enterprise Credit Evaluation

作　　者：吴蔚[1] 僧德文[1] 

机构地区：[1]浙江水利水电专科学校,浙江杭州310018

出　　处：《浙江水利水电专科学校学报》2009年第4期54-57,共4页Journal of Zhejiang Water Conservancy and Hydropower College

基　　金：浙江省水利厅科研基金资助项目(RC0717);浙江水利水电专科学校科研基金资助项目(XK200705);浙江省教育厅科研基金资助项目(Y200805387)

摘　　要：介绍了支持向量机的理论基础和数学模型,阐述了支持向量机在数据分类中的应用和算法描述,提出基于7维以上输入的支持向量机的分类模型,以提高企业信用评价的准确度,并应用MATLAB实现了企业信用评价模型的设计和测试,分别对若干贷款企业进行两类模式分类的研究,信用评价的准确率均达到了100%.结果表明,采用支持向量机对企业信用进行评价是可行的、有效的,为企业信用评价开辟了一个新的途径.This paper stresses on the application of SVM and its algorithmic description on data classification based on the rationale and the mathematical model. The classifiable model based on the 7 dimension input support vectors is also discussed, which upgrades the accuracy of the credit evaluation. The evaluation model for enterprises credit based on MATLAB has been well-designed and well-tested. The research on the two-pattern classification for 106 companies listed on China Stock Exchange by 2000 and 80 borrowers of one domestie commercial bank by the end of 2001 proved the rate of exactness up to 100%. This result indicates its feasiblity and effect in credit rate evaluation of the enterprises, which can become a new method of the credit evaluation of enterprises.

关 键 词：信用评价 财务指标 支持向量机 MATLAB 

分 类 号：O141.4[理学—数学]