摘要
针对现有网络预警技术误报率和漏报率较高、响应时间长的问题,将支持向量机应用到网络安全预警中,并用Python语言进行仿真。首先对原始数据进行预处理,消除不同量纲对模型的影响;其次采用决策树进行特征选择,剔除无用的噪声属性,降低数据维度;然后利用处理好的数据来训练支持向量机模型,进行网络预警检测分类;最后给出仿真结果,并根据分类算法常用评价指标对仿真结果进行评价。结果表明,将支持向量机应用到网络安全预警优于传统预警技术。
Aiming at the problem of existing network security early warning,such as high false alarm rate and missing alarm rate,and long response time,the support vector machine was applied to the network security early warning.Meanwhile Python language was used for simulation.First,the original data was preprocessed to eliminate the influence of different dimensions on the model.Secondly,the decision trees were used for feature selection to eliminate useless noise attributes and reduce data dimensions.Then the support vector machine model was trained with the processed data for intrusion detection classification.Finally,the simulation results were given and the results were evaluated according to the common evaluation indexes of classification algorithms.The results show that the application of support vector machine in network security early warning is better than the traditional intrusion detection technology.
作者
陈丽芳
杨丽敏
于健
CHEN Li-fang;YANG Li-min;YU Jian(College of Science,North China University of Science and Technology,Tangshan Hebei 063210,China)
出处
《华北理工大学学报（自然科学版）》
CAS
2021年第2期132-140,共9页
Journal of North China University of Science and Technology：Natural Science Edition
基金
河北省自然科学基金面上项目(F2014209086)。
关键词
网络安全预警
支持向量机(SVM)
特征选择
分类算法
network security early warning
support vector machine(SVM)
feature selection
classification algorithm
