Abstract-The proliferation of crimes using anonymised networks such as The Onion Router (TOR) has posed considerable hurdles for law enforcement and cybersecurity experts. Conventional forensic methods often have difficulties in tracking illegal activity carried out on TOR because of its encryption and anonymity attributes. This study aims to provide a forensic analysis model tailored for the investigation of criminality inside the TOR network. The model utilises sophisticated data analysis methods, using machine learning classifiers like as Naïve Bayes, Support Vector Machines (SVM), Random Forest, and K-Nearest Neighbours (KNN), to identify anomalous activity and discern attack patterns. Furthermore, it incorporates feature selection techniques to improve classification precision and minimise false positives. The proposed methodology utilises publicly accessible information and network traffic analysis to enhance the detection and investigation of criminality inside the TOR network, providing significant insights for security experts and law enforcement authorities.

Abstract-Cybercrime is a specialised field that use internet communication networks to enhance the identification of cyber offenders via cyber laws. Extensive study is being undertaken to provide appropriate legal methodologies for preventing and regulating cybercriminal activity. Malware and phishing detection have become as prominent subjects in the last decade because to the harm they inflict on internet users. The identification of phishing websites is a novel area in the discipline. Phishing websites are regarded as a significant threat for the exploitation of personal information for the benefit of cybercriminals. This research presents an automated classification system designed to identify malware and phishing websites by integrating several clustering techniques using a cluster ensemble approach. .

DOI: 10.61137/ijsret.vol.10.issue6.425