基于机器学习建模的XSS攻击防范检测

XSS Attack Prevention and Detection Based on Machine Learning Modeling

为了解决网络流量中跨站脚本攻击频发且攻击危害性高的问题,研究了基于机器学习算法建模的跨站脚本检测技术,从复杂的网络流量数据中发掘跨网站脚本(Cross-Site Scripting,XSS)攻击,然后结合专家经验和安全业务知识对数据进行打标学习,并采用机器学习技术训练算法模型,实现了对XSS攻击的自动化和智能检测功能。测试表明,在安全领域引入机器学习算法,能够准确识别复杂多变、高危恶意的XSS攻击,提高了安全设备对威胁攻击的检测能力。

In order to solve the problem of frequent cross-site script attacks and high harmfulness in network traffic, this paper studies the cross-site script detection technology that based on machine learning modeling,explores XSS(Cross-site scripting) attacks from complex network traffic data, then combines with expert experience and security business knowledge to mark and learn the data, and uses machine learning to train algorithm model to realize automatic and intelligent detection of XSS attacks. Experiments indicate that in the field of security, machine learning is introduced to accurately identify complex, changeable and highrisk malicious XSS attacks, which improves the ability of security devices to detect threat attacks.