GraphFVD: Property Graph-based Fine-grained Vulnerability Detection

This is an official implementation of our paper "GraphFVD: Property Graph-based Fine-grained Vulnerability Detection".

Overview

In this repository, you will find a Python implementation of our GraphFVD. As described in our paper, GraphFVD is a novel property graph-based fine-grained vulnerability detection approach. Our approach extracts property graph-based slices from the Code Property Graph and introduces a Hierarchical Attention Graph Convolutional Network to learn graph embeddings.

Requirements

• Python 3.7.2
• Pytorch 1.9.1
• Transformer 4.30.2
• Tokenizers 0.13.3

Training and Evaluation

python my_preprocess.py

This command preprocesses the dataset and splits it into training, validation, and testing sets.

python run.py

This command is used to train and evaluate GraphFVD model.