Advancing Forensic Examination of Cyber Predator Communication Through Machine Learning
Ayodeji Ogundiran, Hongmei Chi, Jie Yan, Ruth Agada · 2024
By analyzing textual and behavioral patterns in online conversations, machine learning algorithms can assist forensic investigators in identifying potential predators, understanding their strategies, and ultimately enhancing the protection of vulnerable individuals. The importance of digital evidence cannot be overstated because digital evidence is often the key to proving the guilt or innocence of someone for actions for which they have been charged when accused of committing a digital crime. The prevalence of criminal activity on social media sites is often very difficult to determine since there are currently no comprehensive statistics on social media crimes. This increased rate of social media crimes has led to an increasing need for social network forensics. Current digital forensic techniques and tools can also not detect patterns efficiently among large datasets from social media and other online digital domains. Artificial intelligence and natural language processing have been seen as solutions to these problems encountered by forensics experts. The research focuses on the development of a robust machine-learning model capable of distinguishing patterns among suspected predatory conversations, thereby offering a proactive approach to combating online sexual exploitation. This study contributes to the ongoing efforts to create safer digital spaces by unveiling the hidden truths embedded in textual content. The findings hold significant implications for developing effective preventive measures, fostering a more secure online environment, and safeguarding individuals from potential harm.