Wenbo Guo
Wenbo Guo
Home
News
Experiences
Publications
Awards
Contact
Blog
CV
Light
Dark
Automatic
Graph Neural Network
HyVulDect: A Hybrid Semantic Vulnerability Mining System Based on Graph Neural Network
We propose a graph neural network vulnerability mining system based on hybrid semantics, which constructs a composite semantic code property graph for code representation based on the causes of vulnerabilities.
Wenbo Guo
,
Yong Fang
,
Haoran Ou
,
Cheng Huang
,
Chun Lin
,
Yongyan Guo
PDF
Vulnerability mining research for open source communities
This project studies the security of source code fragments in open source communities, extracts the core code associated with vulnerabilities through program slicing techniques, and constructs a BiLSTM model for malicious code detection. The text features of posts are also constructed using TF-IDF to train OVO SVMs models for vulnerability code type identification.
PDF
Cite
×