ChamaraVishwajithRajapaksha/Code_Vulnerability_Dataset

---
license: mit
tags:
- security
- cwe
- vulnerability
- code-analysis
- software-security
- dataset
- machine-learning
- llm-finetuning
---

# 🔐 Code Vulnerability Dataset (CWE-Enriched)

## 📌 Overview

This dataset is built from the **bstee615/diversevul** dataset and enhanced with structured vulnerability intelligence from the **MITRE Common Weakness Enumeration (CWE)** database.

It provides a rich, machine-readable representation of software vulnerabilities, mapping raw vulnerable code samples to standardized CWE classifications.

The dataset is designed for research and development in:
- Vulnerability detection models
- Secure code generation
- LLM fine-tuning for cybersecurity tasks
- Static analysis and code understanding systems

---

## 🧠 Dataset Enrichment Process

Each sample in the dataset has been augmented using the MITRE CWE API to include structured security intelligence such as:

- CWE identifier (e.g., CWE-787)
- Vulnerability type (e.g., Out-of-bounds Write)
- Human-readable description
- Severity / exploit likelihood
- Impact categories (e.g., code execution, crash)
- Applicable programming languages
- Security classification metadata

---

## 📊 Data Structure

Each row in the dataset contains:

### 🔹 Original Fields
- `func` → Source code snippet
- `cwe` → Original CWE labels from DiverseVul dataset

### 🔹 Enriched Field
- `cwe_details` → JSON object containing structured CWE metadata:

```json
{
"cwe_id": "CWE-787",
"vulnerability_type": "Out-of-bounds Write",
"description": "The product writes data past the end, or before the beginning, of the intended buffer.",
"severity": "High",
"category": "Memory Corruption",
"impact": [
"Modify Memory",
"Execute Unauthorized Code",
"Crash (DoS)"
],
"languages": ["C", "C++"],
"example": "Example not extracted"
}

---
license: MIT许可证
tags:
- 安全
- CWE
- 漏洞
- 代码分析
- 软件安全
- 数据集
- 机器学习
- 大语言模型（Large Language Model, LLM）微调
---

# 🔐 代码漏洞数据集（CWE增强版）

## 📌 概述

本数据集基于**bstee615/diversevul**数据集构建，并通过来自**MITRE通用弱点枚举（Common Weakness Enumeration, CWE）**数据库的结构化漏洞情报进行增强。

本数据集提供了丰富的机器可读软件漏洞表征，将原始易受攻击代码样本映射至标准化的CWE分类体系。

本数据集面向以下方向的研究与开发：
- 漏洞检测模型
- 安全代码生成
- 面向网络安全任务的大语言模型微调
- 静态分析与代码理解系统

---

## 🧠 数据集增强流程

本数据集的每个样本均通过MITRE CWE API进行增强，以纳入以下结构化安全情报：

- CWE标识符（例如：CWE-787）
- 漏洞类型（例如：越界写入）
- 人类可读的描述
- 严重程度/利用可能性
- 影响类别（例如：代码执行、崩溃）
- 适用编程语言
- 安全分类元数据

---

## 📊 数据结构

数据集中的每一行包含以下内容：

### 🔹 原始字段
- `func` → 源代码片段
- `cwe` → 源自DiverseVul数据集的原始CWE标签

### 🔹 增强字段
- `cwe_details` → 包含结构化CWE元数据的JSON对象：

json
{
"cwe_id": "CWE-787",
"vulnerability_type": "越界写入",
"description": "该产品在目标缓冲区的末端之外或起始之前写入数据。",
"severity": "高",
"category": "内存破坏",
"impact": [
"修改内存",
"执行未授权代码",
"崩溃（拒绝服务）"
],
"languages": ["C", "C++"],
"example": "未提取示例"
}