基于交叉熵和新转移函数的模糊神经网络分类器  

Fuzzy Neural Network Classifier Based on the Cross Entropy Rule and the New Transfer Function

作　　者：毛玲[1] 孙即祥[1] 季虎[1] 

机构地区：[1]国防科技大学电子科学与工程学院,湖南长沙410073

出　　处：《国防科技大学学报》2004年第5期52-56,共5页Journal of National University of Defense Technology

摘　　要：针对目前普遍采用的误差平方和准则及Sigmoid转移函数在BP算法应用中存在的缺陷和不足,提出了基于交叉熵准则和新的S型转移函数构建的模糊神经网络分类器,并将这种分类器应用于心肌梗死的定位诊断,结果表明其训练效率和识别性能都明显优于传统的模糊神经网络。In view of the fact that BP algorithm based on the common used error square sum rule and Sigmoid transfer function have some limitation and shortcomings, the cross entropy rule and a new transfer function are adopted for constructing and training process of the fuzzy neural network classifier. It is used to realize the orientation of myocardial infarction, and the results prove that this classifier has the capability of outperforming the traditional fuzzy neural network in training efficiency and recognizing ability obviously.

关 键 词：交叉熵 转移函数 模糊神经网络分类器 心肌梗死 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]