Detect Darknet URL Based on Artificial Neural Network

Abstract

Darknet is a network that transmits data on the Internet through anonymous network technology and protects the relationship between the two sides of communication from being leaked. Because the IP addresses of both sides of the communication cannot be traced on the darknet, the identity of the user cannot be determined. The darknet is used by criminals to engage in criminal activities. This paper studies the URL address of the darknet, proposes an algorithm for darknet URL recognition using artificial neural network. The algorithm transforms URL into a fixed length vector, and then uses it as a part of the input data of artificial neural network for learning and classification. Experiments show that the proposed algorithm has high accuracy, can accurately identify the darknet URL through multiple iterations under different attribute accuracy. Experimental results show that the proposed algorithm can achieve 99.3% detection accuracy.