Forensic Analysis
Forensic analysis, encompassing the scientific examination of evidence to support legal proceedings, is rapidly evolving with the integration of machine learning and advanced signal processing techniques. Current research focuses on developing explainable AI methods for improved transparency and trust, particularly in cyber forensics and authorship attribution, often employing deep learning architectures like Vision Transformers and Graph Neural Networks, as well as more traditional methods like Support Vector Machines. These advancements enhance the accuracy and reliability of forensic investigations across diverse domains, including image and audio analysis, improving the ability to detect manipulations, identify sources, and ultimately contribute to more just and effective legal outcomes.
Papers
MONet: Multi-scale Overlap Network for Duplication Detection in Biomedical Images
Ekraam Sabir, Soumyaroop Nandi, Wael AbdAlmageed, Prem Natarajan
XG-BoT: An Explainable Deep Graph Neural Network for Botnet Detection and Forensics
Wai Weng Lo, Gayan K. Kulatilleke, Mohanad Sarhan, Siamak Layeghy, Marius Portmann