A Knowledge Graph Question Answering Approach to IoT Forensics

Abstract

Internet of Things (IoT) forensics has been a particularly challenging task for forensic practitioners due to the heterogeneity of IoT environments as well as the complexity and volume of IoT data. With the advent of artificial intelligence, question-answering (QA) systems have emerged as a potential solution for users to access sophisticated forensic knowledge and data. In this light, we present a novel IoT forensics framework that employs knowledge graph question answering (KGQA). Our framework enables investigators to access forensic artifacts and cybersecurity knowledge using natural language questions facilitated by a deep-learning-powered KGQA model. The proposed framework demonstrates high efficacy in answering natural language questions over the experimental IoT forensic knowledge graph.