一种粗糙小波网络分类器及应用

A Rough Wavelet Neural Network Classifier and its Application

提出了一种粗糙小波网络分类器的模型。其过程为：利用粗糙集理论获取分类知识，根据训练样本属性值离散化、属性约简和值约简来构造粗糙小波网络分类器。该分类器可以有效地克服粗糙集规则匹配方法抗噪声能力和规则泛化能力差的缺点；同时可简化小波网络的结构，加快网络的训练速度。并详细介绍了该分类器用于入侵数据识别的步骤和仿真实验结果。

A model of rough wavelet neural network classifiers is proposed.The rough set theory is used to acquire the knowledge of classification,which includes decision table construction, attribute discretization,attribute reduction and attribute＇s value education.The rough wavelet neural network classifier is established according to the reduced attributes,the attribute＇s value education.The proposed classifier has better abilities of antidisturbance and generalization and can simplify the structure of the neural network and speed up the training rate of the network. The steps of applying this classifier to recognition of intrusion data and the results are described in the computer simulation experiments.