摘要
针对软件漏洞挖掘领域的问题,文章探讨了一种基于深度学习的漏洞检测方法。首先,分析了目前常用的漏洞挖掘方法;其次,研究设计了软件漏洞检查的总体框架;再次,采用Word2Vec技术将代码转换为词向量序列,对循环神经网络(RNN)模型进行了训练,从而实现对软件漏洞的自动检测;最后,在TensorFlow框架下构建了检测模型,并利用NVD数据集进行了实验验证。实验结果表明,所提方法在准确率、精确率、召回率和F1-score等指标上表现出良好效果。
Aiming at the problem of software vulnerability mining,this paper discusses a vulnerability detection method based on deep learning.First of all,this paper analyzes the commonly used vulnerability mining methods.Secondly,this research designs the overall framework of software vulnerability detection.Then,this research uses Word2Vec technology to convert the code into word vector sequence,the recurrent neural network(RNN)model is trained to automatically detect software vulnerabilities.Finally,the detection model is built under the TensorFlow framework,and the NVD data set is used to verify in the experiments.The experimental results show that the proposed method performs well in accuracy,accuracy,recall and F1-score.
作者
徐圣林
XU Shenglin(China Pharmaceutical University,Nanjing 211198,China)
出处
《无线互联科技》
2024年第20期95-97,112,共4页
Wireless Internet Science and Technology
关键词
深度学习
循环神经网络
词向量
软件漏洞
deep learning
recurrent neural network
word vector
software vulnerability
