Machine Learning and Artificial Intelligence-based Child Abusing Tracking System for the Detection of Online Sexual Predators

Abstract

With the upward thrust of cybercrime targeting kids, chat logs may be used to hit upon and mark harmful conduct for law enforcement. Children can be helped by this in a great manner from cybercrime. Previously digital forensic investigations were usually done by hand; this traditional approach of relying on the assessment was not reliable. The solution proposed in this paper uses the Digital Forensic Model using machine learning and artificial intelligence-based on various Child Abuse Tracking System supported by Microsoft technology and other companies to facilitate automatic detection of harmful conversations on the chat log. Therefore, the contribution of this paper is to show how the activities in the digital investigation process can be organized to obtain useful results using PhotoDNA technology (integrated CETS) helping law enforcement, fight child pornography when investigating online attackers. In addition, this paper included the study of Artemis, an automated system that scans chats to detect online sexual predators. So far no one has done any study on both of these tools. We have proposed architecture for the detection or efficiently capturing of the predators by enabling advanced technologies like machine learning, artificial intelligence, geographical information systems, and data mining. This architecture is based on the category of child exploitation and the scenario of the integrated model based on the material collected. The collaboration of all these aspects in an updated and efficient manner can come up with an effective result and can help law enforcement to take necessary actions.