摘要
在当前网络安全领域,恶意文件的检测与防御是一个持续的挑战。传统的恶意文件检测技术,如特征提取与匹配、签名检测技术,虽然在特定环境下有效,但面对复杂多变的攻击手段时往往显得力不从心。文章首先介绍了传统的基于机器学习的恶意文件检测方法;接着详细论述了本文设计的智能恶意文件验证系统,分别介绍了系统架构、数据处理方式及神经网络模型的优化等内容;最后通过系统测试,验证了所提方法的有效性。
In the current field of network security,detecting and defending against malicious files is a continuous challenge.Traditional malicious file detection techniques,such as feature extraction and matching,signature detection,although effective in specific environments,often appear inadequate when faced with complex and ever-changing attack methods.The article first introduces traditional machine learning based malicious file detection methods.Then,the intelligent malicious file verification system designed in this article was discussed in detail,including the system architecture,data processing methods,and optimization of neural network models.Finally,the effectiveness of the proposed method was verified through system testing.
作者
卢红宇
夏一鸣
张冰
LU Hongyu;XIA Yiming;ZHANG Bing(State grid Zhejiang Electric Power Co.,Ltd.,Longquan City power supply company,Lishui Zhejiang 323700,China)
出处
《信息与电脑》
2024年第23期119-121,共3页
Information & Computer
基金
国网浙江省电力有限公司丽水供电公司群创项目“国网丽水龙泉公司基于GPT模型的威胁发现与双向阻断技术测试”(项目编号:5211L7240002)。
关键词
改进神经网络
智能恶意文件
系统设计
improving neural network
intelligent malicious files
system design
作者简介
卢红宇,男,本科,助理工程师。研究方向:电气工程及自动化。
