摘要
该文利用神经网络的自学习能力和模糊逻辑的动态性和及时性等特点,将模糊逻辑和神经网络有机地结合起来,构造出了四层模糊神经网络,并用训练神经网络的相应学习算法训练网络,将该模型用于网络业务源特征提取与分类的研究中,并与单纯的神经网络算法相比较。计算机仿真结果表明,模糊神经网络方法比神经网络算法更优越,该文的研究结果为解决网络业务源特征提取与分类奠定了基础。
This paper addresses a four-layer fuzzy neural network system(FNN),which utilizes both the linguistic,hu-man-like reasoning of fuzzy systems and the powerful computing ability of neural networks(NN).The FNN is trained by the Back-propagation algorithms ,which is used to train the NN,and used to study the feature extraction of the network traffic and classifying.Compared with the sole NN,the simulation demonstrates that the FNN not only can classify the network traffic,but also is superior to NN.This paper supplies the fundamental research of classifying network traffic.
出处
《计算机工程与应用》
CSCD
北大核心
2004年第22期3-5,共3页
Computer Engineering and Applications
基金
国家自然科学基金(编号:60374037)资助项目
关键词
模糊神经网络
BP算法
网络业务
特征提取与分类
Fuzzy Neural Network,Back-Propagation algorithms ,network traffic,feature extraction and classifying