火力与指挥控制2025，Vol.50Issue(12)：175-181,7.DOI:10.3969/j.issn.1002-0640.2025.12.023
基于改进模型的PNN的网络流量数据识别分类技术
Research on Recognition and Classification Technique of Network Traffic Data of Probabilistic Neural Network Based on Improved Model
ZHOU Wen 1GAO Yi 1HE Jun 1YAN Jun 1MA Mingxin1
作者信息
- 1. North Automatic Control Technology Institute,Taiyuan 030006,China
- 折叠
摘要
Abstract
Probabilistic neural network(PNN)is used to identify and classify network traffic data,the rapid,efficient and accurate identification of various types of data from massive network data can be realized,and a basis for network attack prevention and network management can be provided.However,the value of smoothing factor σ in the classical PNN model is generally obtained by experience,and its optimization selection method needs to be further explored to improve the performance of PNN model in identifying and classifying network traffic data.A relatively new optimization search algorithm,salp swarm algorithm,proposed in recent years,is adopted to optimize the value of smoothing parameters,and the radial basis function of the PNN model is constructed by using the optimized σ value,thus improving the classical PNN model.On the basis of the public CIC data set and traffic data obtained from public network,the experimental verification is performed on the improved PNN model.By comparing and analyzing the accuracy and time consuming of the improved model and the classical model in classifying the same sample set,It can be seen that the PNN of the improved model has excellent performance in processing network traffic data identification and classification tasks,and its performance has been significantly improved.关键词
概率神经网络/CIC数据集数据/捕获特征/樽海鞘算法/平滑参数/径向基函数Key words
probabilistic neural network/CIC data sets/feature extraction/salp swarm algorithm/smoothing parameter/radial basis function分类
信息技术与安全科学引用本文复制引用
ZHOU Wen,GAO Yi,HE Jun,YAN Jun,MA Mingxin..基于改进模型的PNN的网络流量数据识别分类技术[J].火力与指挥控制,2025,50(12):175-181,7.
