基于图神经网络(GNN)的漏洞检测算法及应用研究
Research on Vulnerability Detection Algorithm and Application Based on Graph Neural Network(GNN)
哈焱1
作者信息
- 1. 蚌埠学院计算机与信息工程学院,安徽 蚌埠 233030
- 折叠
摘要
为提高漏洞检测的准确性,提升软件运行的安全,基于图神经网络(GNN)提出一种漏洞检测算法,并对其具体应用效果进行分析.在此基础上,运用图神经网络(GNN)对特征进行分类,输出漏洞检测的结果.结果表明,提出的漏洞检测算法能更准确地识别真实的漏洞,减少将正常代码错误地标记为漏洞的情况.
Abstract
To improve the accuracy of vulnerability detection and enhance the security of software operation,a vulnerability detection algorithm based on graph neural network(GNN)is proposed,and its specific application effect is analyzed On this basis,graph neural networks(GNNs)are used to classify features and output vulnerability detection results The results indicate that the proposed vulnerability detection algorithm can more accurately identify real vulnerabilities and reduce the situation of mislabeling normal code as vulnerabilities.
关键词
图神经网络/漏洞/检测/算法/模式Key words
graph neural network/loophole/testing/algorithm/pattern引用本文复制引用
出版年
2024
NETL
NSTL
万方数据