Digital Forensics & Artificial Intelligence Research Lab
School of Computer and Cyber Sciences · Augusta University
The Digital Forensics & Artificial Intelligence Research Lab conducts research at the intersection of digital forensics, artificial intelligence, cybersecurity, privacy, mobile systems, and Internet of Things security.
Our mission is to advance AI-based digital forensics and privacy research through collaborative, applied, and reproducible work. DFAIR Lab is directed by Dr. Gokila Dorai and is located at the Georgia Cyber Center in Augusta, Georgia.
DFAIR Lab projects focus on:
- Digital forensics and incident response
- Artificial intelligence and machine learning for cybersecurity
- Internet of Things security and network traffic analysis
- Concept drift, adaptive learning, and intrusion detection
- User data privacy and privacy policy analysis
- Mobile, cloud, and cyber-physical systems security
- Natural language processing for security and privacy applications
| Repository | Description |
|---|---|
XSecIoT |
Machine learning-based intrusion detection for IoT networks, including conformal evaluation, automatic model retraining, multimodal classification, and real-time processing. |
CAPEX |
Framework for generating, capturing, and processing IoT network attack datasets for concept drift evaluation and intrusion detection. |
CADE_FIRCE |
Adaptation of CADE, a concept drift detection and explanation method for security applications, for integration with FIRCE-style streaming evaluation. |
PoliGraphM1 |
Automated privacy policy analysis using knowledge graphs, adapted for M1 MacBooks. |
LabNetworkSim |
Archived digital twin simulation of the lab network using Scapy and NetworkX for packet simulation. |
tcpdump-command-runner |
Archived Python utility for capturing network traffic with tcpdump, running custom commands, and saving PCAP output. |
We develop frameworks for evaluating machine learning-based intrusion detection systems under realistic streaming conditions, including changing network behavior, attack traffic, retraining events, and concept drift.
Relevant repositories:
We study user data privacy risks and develop automated approaches for analyzing privacy policies and privacy-related software behavior.
Relevant repository:
We build tools for capturing, simulating, and processing network traffic to support reproducible cybersecurity and digital forensics research.
Relevant repositories:
DFAIR Lab brings together academia, government, and industry collaborators to study digital forensics, privacy, and AI-driven cybersecurity. We welcome collaboration on research involving:
- AI for digital forensics
- IoT and network security
- Privacy-preserving systems
- Machine learning-based intrusion detection
- Reproducible cybersecurity experimentation
For more information, visit dfairlab.com.
This GitHub organization hosts public research software, datasets, experiments, and supporting tools from DFAIR Lab.
For project-specific questions, please open an issue in the relevant repository.