摘要
传统的恶意软件检测方法依赖人工特征工程,难以应对海量数据和复杂攻击模式,而深度学习通过自动化、高精度和强泛化能力成为恶意软件检测的核心技术。文章提出传统检测方法的局限性、深度学习的革新优势、当前网络安全恶意软件检测存在的问题以及基于深度学习的恶意软件检测模型设计,并从特征提取与表示学习、检测系统架构设计、对抗防御强化等3个方面探讨了深度学习的技术实现路径。研究表明,深度学习通过端到端特征学习和复杂模式挖掘,显著提升了恶意软件检测的准确率和实时性。
Traditional malware detection methods rely on manual feature engineering,which makes it difficult to deal with massive amounts of data and complex attack patterns.However,deep learning has become the core technology for malware detection through automation,high accuracy,and strong generalization ability.The article points out the limitations of traditional detection methods,the innovative advantages of deep learning,the current problems in network security malware detection,and the design of malware detection models based on deep learning.It also explores the technical implementation path of deep learning from 3 aspects:feature extraction and representation learning,detection system architecture design,and adversarial defense reinforcement.Research has shown that deep learning significantly improves the accuracy and real-time performance of malware detection through end-to-end feature learning and complex pattern mining.
作者
李英华
朱景怡
LI Yinghua;ZHU Jingyi(Baoding University,Baoding 071000,China;City University of Macao,Macao 999078,China)
出处
《无线互联科技》
2025年第14期85-88,共4页
Wireless Internet Science and Technology
基金
保定市科技计划项目,项目名称:中小学课后服务中推进编程能力培养的策略研究,项目编号:2440ZZ026。
关键词
深度学习
网络安全
恶意软件
检测
deep learning
network security
malicious software
detection
