🧬 FCG-MFD: Malware Family Detection Dataset

Benchmark dataset using Function Call Graphs (FCG) for malware detection and classification

---

📌 Overview

The FCG-MFD dataset is a large-scale benchmark dataset designed for malware family detection and classification using Function Call Graphs (FCG).

It contains 100,000 samples (50K malware + 50K benign) collected from multiple real-world sources.

This dataset enables advanced research in:

• Malware detection
• Family classification
• Graph-based machine learning
• Cyber threat intelligence

---

📊 Dataset Highlights

• ✅ 100,000 total samples
• ✅ 50,000 malware samples
• ✅ 50,000 benign samples
• ✅ 35+ malware families
• ✅ Function Call Graph (FCG) representation
• ✅ Metadata + behavioral features

---

🧬 Malware Categories

• Trojan
• Ransomware
• Worms
• Backdoor
• Botnet
• Virus

---

🧠 Methodology

The dataset is constructed using:

• Malware sources:

  • VirusShare
  • VirusSample
  • MalwareBazaar
  • VX-Underground
  • theZoo

• Analysis tools:

  • VirusTotal
  • Cuckoo Sandbox

• Feature extraction:

  • Function Call Graphs (FCG)
  • Behavioral analysis
  • Metadata (MD5, PE info)

---

---

🧪 Applications

• Malware detection
• Malware family classification
• Graph Neural Networks (GNN)
• Intrusion Detection Systems (IDS)
• Cybersecurity research

---

---

📜 Citation

If you use this dataset, please cite:

@article{HADI2025104050,
title = {FCG-MFD: Benchmark function call graph-based dataset for malware family detection},
journal = {Journal of Network and Computer Applications},
volume = {233},
pages = {104050},
year = {2025},
issn = {1084-8045},
doi = {https://doi.org/10.1016/j.jnca.2024.104050},
url = {https://www.sciencedirect.com/science/article/pii/S1084804524002273},
author = {Hassan Jalil Hadi and Yue Cao and Sifan Li and Naveed Ahmad and Mohammed Ali Alshara},
keywords = {Malware detection, Malware family classification, Function Call Graph, Dataset},
}

---

👨‍💻 Maintainer

CyberSar Lab 🔗 https://cybersar.kaust.edu.sa/

---

🛡️ Advancing Malware Analysis with Graph-Based Intelligence