The internet has registered exceptional growth over the years. This trend has given rise to the dark web, a somewhat troublemaking facet of the internet that has supported the proliferation of various social ills.
Nonetheless, the entire dark web should not be criminalized. While there are particular positive attributes of the hidden web, the dark web has been notorious for harboring cybercrime, extreme behavior and the drug trade.
This issue has prompted law enforcement agencies to develop innovative strategies to arrest dark web masterminds that have been blamed for the widespread erosion of societies’ moral fiber.
This aspect, therefore, has presented an urgent need for the application of mechanisms used to monitor and analyze dark web activity by observing trends characterizing darknet forums.
Quite unfortunately, it has been evident that the employment of conventional web crawling methods has been ineffective for collecting dark web content.
It is this need that has prompted experts to employ artificial intelligence (AI) techniques in monitoring dark web activity.
Various scientific research papers have sought to explicate the practicality of applying AI as the ultimate dark web cop in combating online crime.
Fundamentally, the Tor network has been extensively studied with the intention of manually labeling crawled sites that have been grouped categorically. Such labelling has been closely associated with the existence of automated crawls.
This perspective has been known to be efficient in creating definite taxonomies that advise on issues surrounding dark web activity.
However, some limitations have been established within the context of specialized law enforcement. Experts have endeavored to monitor illegal activity on the dark web by concentrating on particular forms of crime, such as fraud.
Hitherto, a significant gap exists between taxonomies that explicate dark web activity and those able to provide detailed information on dark web crime.
This challenge has formed the basis of various research questions that seek to streamline law enforcement within the confines of the hidden web.
AI and Dark Web Monitoring
Artificial intelligence has swept the doyens of computer technology off their feet. But what exactly is the significance of artificial intelligence in dark web activity?
This question is, undoubtedly, critical and can be approached by understanding the research milestones made concerning the involvement of AI in controlling the hidden web.
A 2010 scientific paper, A Focused Crawler for Dark Web Forums by Abbasi et al., provided a proposition for the creation of a crawling mechanism that would work to collect content from darknet forums.
The outcomes of the study were very interesting. The researchers were successful in creating a specialized crawler for gathering information from dark web forums.
They employed an accessibility technique that registered monumental levels of success that were recorded at over 90 percent.
The crawler utilized a mechanism that was not tied down by language restrictions. This ranged from URL tokens to anchor text. According to the researchers, this aspect was fundamental in ensuring the gathering of content across the multilingual divide.
The crawler also used particular strategies that would allow it to sustain incremental crawling. Furthermore, the incremental crawling method was combined with a recall improvement technique that would typically re-spider uncollected content.
The paper’s discussion intimated that the study managed to obtain latest information from 109 dark web forums. This success occurred despite the reality of multiple languages spoken in the forum.
Specifically, the forums were grouped into three—American, Middle Eastern and Latin.
Moreover, the researchers provided a case study that vindicated the practicality of the collections in analyzing dark web content.
The case study centered on the particular discussion topics that rule the dark web, and the associated interaction patterns that characterize the respective forums.
In conclusion, the researchers affirmed the premise that a forum crawling system would go a long way in providing an entry into the world of darknet forums with the intention of understanding these covert groups.
In the same sphere, a member of the Australian Federal Police (AFP) intends to invent a crawler that operates on artificial intelligence.
Janis Dalins, the AFP officer, believes that the crawler would be instrumental in the monitoring of dark web activity with the aim of detecting illegal action and alerting law enforcement of the same.
The success of the AI crawler would aid officers in efforts to track down seasoned criminal gangs that engage in all manner of illegal activity.
The different situation is that of a crammed up system where officers spend lots of time to investigate online crime owing to their need, for example, to analyze images in their bajillions.
Dalins intimates that the crawler will contribute to a database maintained by police officers who will monitor dark web activity regularly.
Nonetheless, the officer has declared that the crawler will by no means replace the importance of police officers; it will augment their efforts in maintaining law and order.
Currently, the dark web is mainly accessed through Tor, a highly encrypted network that promises its users the benefit of anonymity as they engage in various online activities. According to Dalins, this aspect has fueled the proliferation of criminal elements on the dark web.
The dark web supports the trade of goods ranging from drugs to stolen records and identities.
To solve this problem, Dalins merged efforts with his Ph.D supervisor to provide the ultimate solution to law enforcement issues within the dark web context. To this effect, their work was published in the Digital Investigation journal.
According to the paper, the researchers operated under permission from Australian authorities. This detail concerned their intentions to carry out an open crawl of Tor for the sole purpose of the study.
This provision also included the go-ahead to analyze illicit content like child exploitation material, which is a rampant form of crime on the dark web.
Just like the previously mentioned scientific paper, Dalins’ work produced intriguing outcomes. The AI crawler managed to collect several hundred thousands of pages from more than 7,000 dark web domains.
This allowed the team to amass information ranging from a broad sphere that heavily leaned on illegal online activity.
The study applied a Tor-use Motivation Model (TMM) to test its hypotheses. TMM is a classification technique that has been specially tailored for police work.
According to the paper, the manual labeling of thousands of unique Tor pages was undertaken, and this unraveled mysteries surrounding the possible applications of AI in curbing dark web crime. The technique exposed networks of money laundering and illegal trade.
The study also discredited the traditional means used by law enforcement in monitoring illicit dark web activity. Concerning this, conventional models proved to be vague within the terms of law enforcement.
Dalins is hopeful that the project will advance as planned. Funding for the second phase will enable the integration of his database with an AI program.
The AI tool will then stalk the dark web and group websites automatically. It would have the ability to alert the AFP whenever it encounters any illegal activity on the dark web.
The Flip Side
Following the above analysis, it would be easy to regard AI as a welcome idea for restoring stability on the dark web.
While this premise might be correct, the application of police-supported AI crawlers on dark web activity may rub many people the wrong way. Notably, this development raises serious privacy concerns.
Privacy advocates would easily dub it as a witch hunt for people out to exercise their freedoms of thought and speech.
While criminals have primarily misused the dark web, a significant portion of the hidden web has been instrumental in circumventing government-funded media censorship and being a space where whistleblowers can expose corruption anonymously. This is just but one positive attribute borne in the dark web that is often undermined.
Allowing the police to advance their surveillance methods on the dark web will heighten issues surrounding political oppression. In the subject of child exploitation, such a move may be justified.
However, the privacy and security of individuals using the dark web for legitimate reasons will become primarily compromised.
It has always been important for people to have complete control over their online privacy, with an emphasis on the prevention of government spying on their online lives—privacy is a human right.
The application of AI on monitoring dark web activity has both academic and scientific value. The reflection of collected data on security informatics would go a long way to combat dark web crime.
However, there are serious privacy concerns attached to the application of AI in checking dark web sites.
As a bona fide citizenry, it would be prudent for people to draw the line between constructive dark web surveillance and unwarranted government interference on activities of the “great hidden web.”
Latest posts by Richard (see all)
- Reddit Hit by Security Breach, User Data Exposed - August 8, 2018
- Murder-For-Hire Charge Dismissed Against Ross Ulbricht - August 7, 2018
- Supreme Court Denied Ross Ulbricht’s Appeal, Supporters Launches New Petition - July 24, 2018