基于僵尸网络流量特征的深度学习检测研究  

Research on Deep Learning Detection Based on Botnet Traffic Features

在线阅读下载全文

作  者:王亚东 WANG Yadong(Shandong Vocational College Shandong,Jinan 250104)

机构地区:[1]山东职业学院,山东济南250104

出  处:《长江信息通信》2025年第2期106-108,共3页Changjiang Information & Communications

摘  要:为提高检测僵尸网络攻击的能力,加强网络安全防护,首先概述了僵尸网络检测流程,重点探讨了特征提取方法,包括访问URL的频率、不同URL的个数、访问顺序、不同URL长度的个数、无UserAgent的访问个数以及访问URL的不同时间间隔个数等六个关键特征。经过归一化处理后,采用多层感知器构建深度学习检测模型,并进行模型训练。检测结果分析与讨论表明,该模型准确率约为65%,六个特征值共同作用可较准确、全面地描述僵尸网络流量,有效提高了检测模型的准确率。in order to improve the ability of detection botnet attack,strengthen network security protection,first summarizes the botnet detection process,mainly discusses the feature extraction method,including the frequency of access URL,the number of different URL,access order,the number of different URL length,no UserAgent number of access and access the URL different time interval number of six key features.After normalization,a multilayer perceptron is used to construct a deep learning detection model and perform model training.The analysis and discussion of the detection results show that the accuracy of the model is about 65%,and the six eigenvalues can accurately and comprehensively describe the botnet traffic,which can effectively improve the accuracy of the detection model.

关 键 词:僵尸网络 深度学习 检测模型 

分 类 号:TP391[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象