摘要
结合自然语言处理技术,采用卷积神经网络算法训练SQL注入检测模型,主要包括文本分词处理、提取文本向量和训练检测模型三个部分。实验结果与BP神经网络算法结果对比,发现基于卷积神经网络的SQL注入检测模型仅需提取用户输入的信息,就可以对攻击行为进行检测,具有很强的预测能力,同时针对变异SQL注入攻击具有良好的识别能力。
This paper combines natural language processing technology and uses convolution neural network algorithm to train SQL injection detection model.It includes three parts:text segmentation processing,extracting text vectors and training detection models.By comparing the BP neural network algorithm,the experimental results show that the SQL injection detection model based on the convolution neural network only needs to extract the information from the user input,and can detect the attack behavior,which has a strong prediction ability and is good for the variant SQL injection attack with clockwise.At the same time,it has good recognition ability against variant SQL injection attacks.
作者
叶永辉
谢加良
李青岩
YE Yonghui;XIE Jialiang;LI Qingyan(College of Science,Jimei University,Xiamen 361021,China)
出处
《集美大学学报(自然科学版)》
CAS
2019年第3期234-240,共7页
Journal of Jimei University:Natural Science
基金
国家自然科学基金资助项目(11371130)
福建省自然科学基金资助项目(2017J01558)
福建省中青年教师教育科研项目(JAT160696,JA15265)