ASIC: Web scammers should watch their language

A new Internet document classifications system which employs linguistics theory to identify online breaches of the Corporations Act is set to become the latest weapon in the Australian Securities and Investments Commission's (ASIC) arsenal against Australian Web scammers.

Dubbed "Scamseek", the new technology claims to have the potential to assess and aggregate the risk associated with information on a Web site using a combination of language analysis and Web scanning technology.

Scamseek will be developed by ASIC through a new joint research initiative with the Capital Markets Cooperative Research Centre (CMCRC), Sydney University and Macquarie University. The project has an overall budget of just over $1 million.

In a statement announcing the new research project, team leader for the CMCRC and the University of Sydney professor Jon Patrick said the system will use new theories on textual meanings "to unravel the deep linguistic features that will enable us to detect scam proposals no matter what surface form of language they use".

As well as utilising language technology, the project will also apply a specialist ‘Web spider' to search out potential Web sites, using technology developed by CMCRC member Security Markets Automated Research Training and Surveillance (SMARTS).

ASIC director of electronic enforcement Keith Inman said while the linguistics technology behind Scamseek is new, ASIC has been conducting scans for potentially suspicious Web sites for several years.

Initially, scans were carried out through routine "surf days". These involved scouting the Internet for potentially suspicious sites using ASIC's specially grafted list of scam keywords. Although the theory behind the surf days was sound, Inman said the team's success rate was low: for every one or two suspicious sites found, 1000 legitimate sites were also being analysed.

In a bid to hone down the ratio of legitimate versus suspicious sites, the team developed an automatic tool which scanned sites using ASIC's keywords list. The process however, was still too "labour intensive", Inman said.

Unlike ASIC's current methods, the Scamseek tool will elaborate on the keywords list by devising a "style sheet", analysing not just the words used on an offer but the nature and relationship of the terms throughout the offer itself, he said. Because the Scamseek tool will be automatic, it will also allow ASIC to scan the Internet 24 hours a day.

"The system we are developing through this joint project will be an eye that never sleeps, constantly seeking out sites that we can take action against," Inman said.

The first phase of the research project is expected to take six months. If fruitful, the CMCRC will launch a commercial version of the tool.

Inman said based on the preliminary work already done by the University of Sydney's linguistic department, ASIC is confident the tool will become an integral part of their crack down on Internet scams.

Although Scamseek will help to identify potential fraudsters, Inman also urged those looking for more information on ASIC or wanting to lodge a complaint to visit ASIC's Web site:

Join the newsletter!


Sign up to gain exclusive access to email subscriptions, event invitations, competitions, giveaways, and much more.

Membership is free, and your security and privacy remain protected. View our privacy policy before signing up.

Error: Please check your email address.

More about Australian Securities & Investment CommissionMacquarie UniversityMacquarie UniversityUniversity of SydneyUniversity of Sydney

Show Comments