Register Now


Lost Password

Lost your password? Please enter your email address. You will receive a link and will create a new password via email.

Add post

You must login to add post .

Add question

You must login to ask a question.


Register Now

Welcome to! Your registration will grant you access to using more features of this platform. You can ask questions, make contributions or provide answers, view profiles of other users and lots more. Register now!

Machine with power to detect hackers BEFORE they have committed a crime tested by Cambridge scientists

A machine with the power to identify cyber-criminals who may be plotting to commit an offence has been tested successfully by experts. Scientists at Cambridge University have tested a system likened to George Orwell’s “Thought Police” for scanning the web to identify potential cyber criminals based on how they were commenting in forums.

But the researchers who piloted the programme denied it would be used by police forces to lock people up before they commit a crime, like in the Tom Cruise movie and dystopian story by Philip K. Dick, “Minority Report”.

Computer scientists used machine learning to crunch swathes of data from 113 known cyber-attackers. They built algorithms to compare the data to thousands of underground forum users, and scanned comments for the warning signs that they may be planning a cyber-attack.

The findings come as the need for methods to prevent cyber attacks is growing. This week it emerged that Vladimir Putin’s spies had attempted to hack the international watchdog investigating the Salisbury nerve agent poisonings, Foreign Office computers in Whitehall and defence laboratories in Porton Down.

The cast of the 2016 of Orwell’s 1984, by American Repertory Theater came to London’s West End. (Manuel Harlan)

They whittled the accounts down to 80 individuals who were highly likely to become an “actor” in a cyber-attack, and when the team went back to read the comments first hand, the researchers said it was clear there was certainly cause for suspicion.

The computer “identified variables relating to forum activity that predict the likelihood a user will become an actor of interest to law enforcement,and would therefore benefit the most from intervention,” said the published paper.

“This work provides the first step towards identifying ways to deter the involvement of young people away from a career in cybercrime,” the paper said.

Dr Alice Hutchings worked alongside the Cambridge Cybercrime Centre at the university’s Department of Computer Science and Technology on the research.

She said the team looked at a large pool of a quarter of million users and believe the system could be used up by cyber-crime police as a way of detecting “risky” individuals.

The technique worked by processing some 30 million posts from the Hackforums website, looking for key words and references to criminal activity, such as “DDoS” referring to a Denial of Service attack, or people who discussed distributing malware and “account cracking”.

“The National Crime Agency does have a preventative strategy within their cyber-crime unit,” she added. “They’ve said they want to be able to divert people away from serious activity.”

But she dreads the idea that it could be used by so-called “Thought Police” to lock-up potential criminals pre-emptively.

The New Scientist made the comparison between the Cambridge experiment and the authoritarian police in George Orwell’s 1984, in which “Thinkpol” officers seek out “Thoughtcrimes” – punishing people for believing anything that goes against the government. But it could perhaps play the role of a “Thought Social Worker”, using warning signs to intervene before young people turn into criminals, the scientist said.

Dr Hutchings said: “I deplore the idea of thought police arresting people before they commit crimes.

“You shouldn’t be held liable for something you’ve just thought of doing, or spoken about but we need some kind of system to intervene when someone is at risk.”

“The aim of doing this is to be able to divert people away from the criminal justice system, by identifying who is most at risk of being prosecuted and putting them in a pro-social pathway. I don’t want young people to be arrested, I want to see a successful intervention.

“Young people are being drawn to this kind of activity, and they are often very talented and intelligent. When they end up in the criminal justice system it is very stigmatising – they end up with fewer prospects, and it can ruin their entire lives.”

How ‘machine learning’ discovered network of hack-attackers

The study hoped to learn more about the “criminal pathways” for young people who use the underground chat site, Hackforums, and discovered users who were performing DDoS (denial of service attacks), Remote Access Trojans, distributing malware, “account cracking”, and operating “bot shops”.And by looking at common behaviour patterns they were able to predict the paths of activity that users would take – such as by looking at who they associate with, and where they learn the skills to commit complex crimes.The team used Natural Language Processing to scan 30 million posts, and were able to pinpoint technical jargon and online idiosyncrasies which suggested criminal intentions. They combined this with data on the users’ popularity on the forum, how they moved from gaming related posts to hacking posts, and how some moved from asking users on the forum for help – to giving help to others.The research identified a particular network of closely connected “actors” exhibiting signs that they could be plotting serious attacks, and to clarify the discovery the scientists read their posts to confirm they were discussing the spreading of malware.The paper said in its conclusion: “We have developed tools for detection and prediction of actors involved in cybercrime activities. These tools help to identify user accounts that might require further investigation by law enforcement and security firms monitoring underground communities.”

Cyber crimes have had devastating consequences in the past and can sometimes by driven by tech-minded youngsters.

One young hacker from Hertfordshire created a programme that fuelled more than 1.7 million attacks last year causing millions in damage when he was only 15.

His “TitaniumStresser” code allowed customers (he charged a membership of £250) to disrupt any website they liked, causing immeasurable losses to thousands of individuals, businesses and other organisations.

The need to prevent cyber-attacks is incredibly pressing today.

Government officials this week accused Russia of conducting a Blitzkrieg of attacks, against chemicals weapons watchdogs in the UK, US and Netherlands – allegedly by the Kremlin’s own “Sandworm” hacking unit.

Four Russian men with diplomatic passports were arrested in the Netherlands after an attempted attack on the OPCW laboratories, which are aiding the UK in investigating the nerve agent attack on Sergei Skripal and his daughter Yulia.

The team were attempting a “brazen” close-range hack into the facility’s systems, and it followed a previous attack against Porton Down – one of the UK’s most secretive military research centres.




About Marie

Leave a reply