Engineering faculty researchers at Ben-Gurion University of the Negev reported yesterday that they have developed a system that can identify 95 percent of Internet pages with terrorism-related content.

The experimental system, which is being developed to detect information regarding terror activity automatically, was designed by Dr. Mark Last of the Department of Systems Information Engineering at BGU, and Prof. Abraham Kandel of the National Institute for Systems Test and Productivity, in the United States.

The system is based on the recognition of patterns in texts with terror content, based on examples from existing Internet sites. It uses these patterns to identify "hits" by surfers on other sites with similar characteristics, in order to locate users affiliated with terror organizations and new sites set up by terrorist elements, among other things.

According to Last, the development has great importance in view of the considerable use of the Internet in coordinating and orchestrating terror acts.

"The lack of ability to enforce limitations on Internet users allows terror organizations to set up Internet sites that spread incitement, raise money in support of terror and find new supporters for their causes," Last said.