A Link Analysis Algorithm for Identification of Key Hidden Services

Abstract

The Tor dark web network has been reported to provide a breeding ground for criminals and fraudsters who are exploiting the vulnerabilities in the network to carry out illicit and unethical activities. The network has unfortunately become a means to perpetuate crimes like illegal drugs and firearm trafficking, violence and terrorist activities among others. The government and law enforcement agencies are working relentlessly to control the misuse of Tor network. This is a study in the similar league, with an attempt to suggest a link-based ranking technique to rank and identify the influential hidden services in the Tor dark web. The proposed method considers the extent of connectivity to the surface web services and values of the centrality metrics of a hidden service in the web graph for ranking. The modified PageRank algorithm is used to obtain the overall rankings of the hidden services in the dataset. Several graph metrics were used to evaluate the effectiveness of the proposed technique with other commonly known ranking procedures in literature. The proposed ranking technique is shown to produce good results in identifying the influential domains in the tor network.