Classifying Darknet Traffic Using Machine Learning to Identify Tor and Suspicious Network Activity
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Through the use of several encrypted relays, darknet networks like Tor allow users to remain anonymous.Although these networks provide anonymity, undesirable activities like cybercrime, illicit data exchange, and anonymous attacks routinely take advantage of them.As a result, identifying darknet traffic has grown in importance for contemporary cybersecurity systems.Complex traffic patterns produced by Tor and other anonymisation services are frequently missed by conventional rule-based intrusion detection systems.In order to identify and categorise network flows into Normal Traffic, Suspicious Traffic, and Onion (Tor) Traffic, this study suggests a machine learning-based darknet traffic analysis system.To examine traffic behaviour, the system makes use of network flow characteristics such as Port Number, Packet Size, Flow Duration, and Packet Count.The machine learning model is trained using a dataset of labelled network traffic records.In order to enhance classification performance, irrelevant attributes are eliminated and specific features are employed during preprocessing.Users can enter traffic parameters and view real-time network behaviour by integrating the trained model into a web application created with Flask and Python.In addition to providing danger information and suggested solutions, the system automatically anticipates the type of traffic.The suggested method can successfully detect Tor-related traffic patterns and differentiate them from typical network activity, according to experimental data.By offering a clever and intuitive traffic classification platform, the suggested solution enhances cybersecurity analysis and darknet traffic monitoring.Real-time packet capture, deep learning methods, and integration with extensive network monitoring systems are possible future enhancements.
Fields of Study & Subjects
Internet Traffic Analysis and Secure E-voting
Network Security and Intrusion Detection
Digital and Cyber Forensics
Publication details
- Journal / source
- International Research Journal of Modernization in Engineering Technology and Science
- Publication date
- 27 March 2026
- Language
- EN