CAJ | 학술논문

网络中大量的恶意网页已经成为网络用户的主要安全威胁。本文提出了一种基于机器学习分类器的网页恶意JavaScript代码分析方法。通过对训练样本训练学习，建立分类模型，最后对测试样本检测。实验表明，本方法能够有效的检测出大部分恶意网页JavaScript代码，检测准确率达到88．5％
망락중대량적악의망혈이경성위망락용호적주요안전위협。본문제출료일충기우궤기학습분류기적망혈악의JavaScript대마분석방법。통과대훈련양본훈련학습，건립분류모형，최후대측시양본검측。실험표명，본방법능구유효적검측출대부분악의망혈JavaScript대마，검측준학솔체도88．5％
A large number of malicious website in the network has become a major security threat of network users. This paper puts forward a kind of malicious JavaScript code analysis method based on machine learning classifier. Through the study of the training sample training, establish the classification model, finally, the detectin is tested on soonples. The experimental results show that this method can more effectively detect the most malicious JavaScript code, accuracy up to 88.5 %.