Content-based technical solution for cyberstalking detection

Abstract

The continued usage of technology has led to the rise of cyberstalking. Cyberstalking is seen as traditional method of stalking that has been altered by technology. This crime has now been modernized using technological tools and techniques. The continued increase in cyberstalking in the world today has drawn attention to the need to address this problem. Though studies on this crime have been conducted in the fields of criminology, legal, public health, sociology, and psychology, it still remains a challenge to detect, prevent, and investigate this crime. Traditional stalking methods have been used to combat it, despite the fact that this crime is committed online. Unfortunately, these methods have provided few solutions for detecting and preventing it. The prevalence of this crime, combined with technological advancement, has necessitated the development of technical strategies to mitigate it, protect victims, and assist law enforcement agencies. In this study, a content-based detection framework for cyberstalking is proposed. The framework consists of message identification, filtering, detection (content detection and profiling offender) and evidence modules. It is designed as a forensic readiness framework that can automatically detect cyberstalking, gather evidence and profile potential offenders. The framework employs machine learning, data mining techniques, digital forensics, and profiling to analyze text, image, and media contents, collect evidence, and profile offenders. This framework would not only detect cyberstalking automatically, but it would also be useful as an investigative tool for law enforcement.