NVD Feature Analysis

NVD Feature Analysis

Analyse the features of NVD vulnerability reports in terms of CWE, CAPEC, threat and CVSS V2/V3.

Table of Contents

• About the Tool
• Generate your own data
• Directly use historical datasets
• Some data visualisation
• Cite

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:

• Report with Threat labels:

• Report with CVSS Version 2 labels:

• Report with CVSS Version 3 labels:

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.

• CWE/CAPEC Labelled Vulnerability dataset (2002-2021) with CWE labels and names, and CAPEC labels.
• Threat Labelled Vulnerability dataset (2002-2021) with labels assigned by cvedetails.
• CVSS V2 Labelled Vulnerability dataset (2002-2021).
• CVSS V3 Labelled Vulnerability dataset (2002-2021).

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:

• Threat distribution:

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},
}