Honghe Zhou, Towson University; Weifeng Xu, University of Baltimore; Josh Dehlinger, Suranjan Chakraborty, and Lin Deng, Towson University
We have developed an automated approach for gaining criminal insights with digital evidence networks. This thrust will harness Large Language Models (LLMs) to learn patterns and relationships within forensic artifacts, automatically constructing Forensic Intelligence Graphs (FIGs). These FIGs will graphically represent evidence entities and their interrelations as extracted from mobile devices, while also providing an intelligence-driven approach to the analysis of forensic data. Our preliminary empirical study indicates that the LLM-reconstructed FIG can reveal all suspects' scenarios, achieving 91.67% coverage of evidence entities and 93.75% coverage of evidence relationships for a given Android device.
Open Access Media
USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.
