摘要： 分析一般模糊极大-极小神经网络的基本原理，阐述模糊计算方法在分类中的准确性和高效性。将一般模糊极大-极小神经网络应用于企业资信评估中，实现模糊区间的输入，缩小企业评估指标定量化中的误差范围。资信评估结果表明，该算法能快速、有效地对企业进行分类，为资信评估提供了解决方案。

关键词: 资信评估, 模糊集, 一般模糊极大-极小神经网络, 超盒, 隶属函数

Abstract: By analyzing the basic principles of General Fuzzy Min-Max(GFMM) neural network and the accuracy and high performance of fuzzy computation for information intelligent processing, the GFMM neural network is applied to the corporation’s credit rating. With genuine inputs of fuzzy realized, the quantitative inaccuracy of standards of evaluating corporations is alleviated to a large degree. Through credit rating towards companies, it is proved that the algorithm can classify companies availably at a high speed. A new project to credit rating is proposed.

Key words: credit rating, fuzzy set, General Fuzzy Min-Max(GFMM) neural network, hyper box, membership function

中图分类号: 

• TP18