SkillAgentSearch skills...

NVDFeatureAnalysis

Correlate NVD datasets wIth CWE/CAPEC/CVSS labels for customised usage. Plus static analysis and data visualisation.

Install / Use

/learn @Yuning-J/NVDFeatureAnalysis
About this skill

Quality Score

0/100

Supported Platforms

Universal

README

NVD Feature Analysis

<!-- PROJECT LOGO --> <br /> <p align="center"> <a href="https://github.com/Yuning-J/NVDFeatureAnalysis"> </a> <br /> <!-- Badges --> <img src="https://img.shields.io/github/repo-size/Yuning-J/NVDFeatureAnalysis?style=for-the-badge" alt="GitHub repo size" height="25"> <img src="https://img.shields.io/github/last-commit/Yuning-J/NVDFeatureAnalysis?style=for-the-badge" alt="GitHub last commit" height="25"> <img src="https://img.shields.io/github/license/Yuning-J/NVDFeatureAnalysis?style=for-the-badge" alt="License" height="25"> <br /> <h3 align="center">NVD Feature Analysis</h3> <p align="center"> Analyse the features of NVD vulnerability reports in terms of CWE, CAPEC, threat and CVSS V2/V3. </p> </p> <!-- TABLE OF CONTENTS -->

Table of Contents

About the Tool

Some static analysis and visualisation for vulnerability instances disclosed in NVD. This tool also generates several datasets with various labels (CWE labels, CAPEC labels, CVSS labels, cvedetails labels) that can be used for further text-mining usage. You can directly employ the generated dataset, or apply this tool to get your own.

Some data samples are illustrated below.

  • Report with CWE and CAPEC labels:
<p align="center"> <img src="https://github.com/Yuning-J/NVDFeatureAnalysis/blob/main/labelledDataset/samplePic/dataSample_cwecapec.png" alt="System" width="800px"> </p>
  • Report with Threat labels:
<p align="center"> <img src="https://github.com/Yuning-J/NVDFeatureAnalysis/blob/main/labelledDataset/samplePic/dataSample_threat.png" alt="System" width="850px"> </p>
  • Report with CVSS Version 2 labels:
<p align="center"> <img src="https://github.com/Yuning-J/NVDFeatureAnalysis/blob/main/labelledDataset/samplePic/dataSample_cvssv2.png" alt="System" width="800px"> </p>
  • Report with CVSS Version 3 labels:
<p align="center"> <img src="https://github.com/Yuning-J/NVDFeatureAnalysis/blob/main/labelledDataset/samplePic/dataSample_cvssv3.png" alt="System" width="800px"> </p>

Generate your own data

Play with vulnerability reports as you want.

  • Step 1: Clone the repo using the following command:
git clone https://github.com/Yuni0217/NVDdata-analysis-and-visualisation.git 
  • Step 2: Create a virtual environment, also change the absolute path in the config.py file.

  • Step 3: Install requirements using pip:

pip install -r requirements.txt
  • Step 4: Download datasets from NVD feeds and assign CWE/CVSS/threat labels.
python src/getNVDdata.py that downlads datasets from NVD feeds.
python src/assignThreatLabels.py that assigns threat labels to NVD vulnerabilities.
python src/assignCWECAPECLabels.py that assigns CWE and CAPEC labels to NVD vulnerabilities.
python src/assignCVSSLabels.py that assigns threat labels to NVD vulnerabilities.

Directly use historical datasets

Note that these datasets are compressed due to large sizes.

Some data visualisation

Check some of the visualisation documentations in the Notebooks. Or create your own awesomes. Below are some of the examples:

  • Vulnerability distribution in terms of access vector:
<p align="center"> <img src="https://github.com/Yuning-J/NVDFeatureAnalysis/blob/main/labelledDataset/samplePic/V2AccessVectorDistribution.png" alt="System" width="700px"> </p>
  • Threat distribution:
<p align="center"> <img src="https://github.com/Yuning-J/NVDFeatureAnalysis/blob/main/labelledDataset/samplePic/CVEDetails2021.png" alt="System" width="780px"> </p>

Cite

If you use this tool in your academic work you can cite it using

@Misc{nvdfeatureanalysis,
  author       = {Yuning Jiang},
  howpublished = {GitHub},
  month        = jan,
  title        = {{NVD Feature Analysis}},
  year         = {2022},
  url          = {https://github.com/Yuning-J/NVDFeatureAnalysis},
}
View on GitHub
GitHub Stars13
CategoryDevelopment
Updated1y ago
Forks5

Languages

Jupyter Notebook

Security Score

80/100

Audited on Jan 6, 2025

No findings