基于模糊识别模型的信用风险监测研究.pdfVIP

第26 卷 第12 期 计算机工程与设计 2005 年12 月 V ol . 26 N o . 12 Computer Engineering and Design Dec. 2005 基于模糊识别模型的信用风险监测研究 1,2 2,3 范艳峰 ， 甄 彤 (1. 西北工业大学计算机学院，陕西西安710072 ；2. 河南工业大学信息科学与工程学院，河南郑州450052 ； 3. 华中科技大学控制科学与工程系，湖北武汉430074) 摘 要：企业资信是一个典型的模糊性问题，利用模糊数学中的多层次系统模糊综合评判法作为企业资信评估的主要 方法，对企业的资信做出全面、综合、系统的评价。提出了基于隶属函数和贴近度的识别方法评估企业中长期贷款偿 还能力，阐述了评估指标标准库的建立和隶属函数的构造。通过实例进行了验证，拓宽了模糊识别理论的应用范围。 关键词：模糊识别 隶属函数 贴近度 信用风险 特征函数 ; ; ; ; 中图法分类号： 文献标识码： 文章编号： ( ) TP39 1 A 1000-7024 2005 12-32 16-03 Research of credit risk detection based on fuzzy pattern recognition FAN Yan-feng1,2, ZHEN Tong2,3 (1. College of Computer, Northwestern Polytechnical University, Xian 710072, China; 2. College of Information Science and Engineering, Henan University of Technology, Zhengzhou 450052, China; 3. Department of Control Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, China) ： Abstract Enterprise asset credit is a kind of typical fuzzy pattern problem. A technology based on multi-layer system fuzzy synthetical evaluation was used for the credit in enterprise. A method based on membership function and approximation value was utilized to evaluate the paying ability for middle term load, and the construction of the evaluation index system and the membership function was presented. It is proved by some examples and extended the application of fuzzy pattern recognition. ： Key words fuzzy recognition ; membership function; approximation value; credit risk