Source Code Vulnerability Detection Project

Summary

This project was aimed at studying and implementing various researchers' methodologies of detecting vulnerabilities within source code. Five of the methods that were studied were also implemented. Each is located within a separate directory in this project. None of the software in this repository is original to this project. It belongs to the corresponding cited authors within the READMEs of each directory.

Running this project

1. Clone the repo

   git clone https://github.com/DanielJRamirez/source-code-vulnerability-detection-project.git

2. Enter the directory of the desired code implementation

   cd target_directory_here

3. Follow the instructions found in the target directory's README

Method/Directory Legend

Method: "On using distributed representations of source code for the detection of C security vulnerabilities"
Directory: code2vec-implementation

Method: "Devign: Effective Vulnerability Identification by Learning Comprehensive Program Semantics via Graph Neural Networks"
Directory: devign_implementation

Method: "Automated Vulnerability Detection in Source Code Using Deep Representation Learning"
Directory: svd_exp1

Method: "How to Detect Software Vulnerabilities in Source Code Using Machine Learning"
Directory: Master-thesis-main

Method: "VUDENC: Vulnerability Detection with Deep Learning on a Natural Codebase for Python"
Directory: vudenc