MIT Developed New Cyber Crime Detection System that Produces Results with 85% Accuracy

A team of researchers from Computer Science and Artificial Intelligence Laboratory (CSAIL), MIT, together with startup PatternEx has developed an artificial intelligence system, AI2, which is the best so far at detecting cyber-attack.

What makes this tool unique is the use of human intuition and machine learning abilities. It has been tested on 3.6 billion log lines triggered over a period of three months by millions of users. This new system is thrice more accurate in detecting the cyber attack.

The systems used previously were either virtual-machine based systems of exclusively operated by humans. Individually, they have not been able to detect the cyber crime like they should. However, the new system uses traits of both the systems to produce something more efficient so that it can make up for their shortcomings.

"This paper brings together the strengths of analyst intuition and machine learning and ultimately drives down both false negatives and false positives. This research has the potential to become a line of defense against attacks such as fraud, service abuse and account takeover", said Nitesh Chawla from University of Notre Dame.

A recent report also stated, "Cyber attack spotters work in two main ways. Some are AI that simply looks out for anomalies in internet traffic. They work, but often throw up false positives-warnings about a threat when actually nothing's wrong. Other software systems are built on rules developed by humans, but it's hard to create systems like that which catches every attack."

This new tool, according to the researchers, can work with 85% accuracy which is three times better than the previous tools. Moreover, it can significantly reduce the number of 'false positives,' an event that is incorrectly identified as a threat, by a factor of five.