Cognitive Cyber Crimes in the Era of Artificial Intelligence

20Investigative Approaches for Dark Web Cognitive Cybercrimes

Anjali Rawat¹* and Anand Rajavat²

¹Computer Science Department, Shri Vaishnav Vidyapeeth Vishwavidyalaya, Indore, India

²Department of Computer Science Engineering, Director, Shri Vaishnav Institute of Information Technology, Shri Vaishnav Vidyapeeth Vishwavidyalaya, Indore, India

Abstract

The proliferation of cognitive cybercrimes on the dark web necessitates advanced investigative methodologies. This study introduces a novel approach using a modified stacking ensemble learning algorithm to detect and classify illicit activities within dark web forums. Utilizing the Darkoob dataset, comprising over 100,000 labeled instances of dark web activities, our method integrates long shortterm memory networks, BERT (Bidirectional Encoder Representations From Transformers), and random forest classifiers to enhance detection accuracy. The proposed model achieved an accuracy of 96.74%, precision of 93.16%, recall of 93.48%, and an F1 score of 96.74%, surpassing existing models in performance metrics. This approach demonstrates significant potential in augmenting cyber threat intelligence and law enforcement capabilities.

Keywords: Dark web, cognitive cybercrimes, ensemble learning, LSTM, BERT

20.1 Introduction

The dark web [1, 2] serves as a clandestine platform facilitating various cognitive cybercrimes, including identity theft, financial fraud, and the dissemination of illicit materials. Traditional investigative techniques ...