Team Develops Accurate Prototype to Detect Fake Sites

Despite the wide use of the internet, there is a surprising lack of awareness amongst most users about fake sites on the net. The number of sites is on the rise and the number of victims is also rising at an accelerating pace. Now a team of researchers from the Arizona Eller College of Management have developed a prototype system that can detect fake websites

The system can detect fake sites much better than a human can. The article by the team on this development got published in the prestigious journal MIS Quarterly, published by the Management Information Systems Research Center, Carlson School of Management and the University of Minnesota. An MISQ publication is a trophy in one’s research career.

Importance

The importance can hardly be over-emphasised. Fake sites are responsible for robbing people of a lot of money or phishing attempts and other such acts of cyber crimes.

Ahmed Abbasi, the lead author of the paper and now a University of Virginia professor of information technology, says:

The problem we’re looking at is quite big. Fake websites constitute much of the Internet fraud’s multi-billion dollar industry, and that is monetary loss…we can’t even quantify the social ramifications. That’s the whole motivation. It is so profitable for fraudsters, and it is slipping through the cracks.

The Methodology

The methodology was two-part.

The first involved checking the easily verifiable information. The prototype checked whether a site’s URL contained ‘http’ instead of an expected ‘https’. Further, it can check when the site was last updated, whether a security key is missing or if the images appear unexpectedly pixelated.

The second part used more hidden information like URL length, number of links and to which sites it is linking to, the character types on the site and how the FAQ section on a site is maintained.

Way forward

Abbasi emphasises the use of great amounts of information for improving the prototype. Fake sites are constantly evolving their look and how they look is not really a criteria any longer, which makes it all the more difficult for a human to detect them. The days of long-held idea of a fake site having a lot of Unicode characters and other unreadable nonsense are gone.

The project is still underway and very much on track. It is, however, a long way from sneaking into our own lives and saving us from a lot of harassment.