The search engine technology of the future finds "hidden" information

At the University at Buffalo researchers have recently finalized a new search technology that is to find "hidden" information on regular web sites.

This time we are not talking about information that is not available to search engine spiders. Rather, this is a project attempting to interpret the information found in an new and more productive way.

The technology is called Unintended Information Revelation or UIR. The press release on this research has been given the title In War on Terrorism, New Search Engine Seeks Hidden Vulnerabilities.

When developing the prototype the researchers have made an experiment, trying to get out information from the American 9/11 Commission Report that can not be found by using regular search engines.

They tested the relationship between three terms that they knew had a connection: "Hamburg," "San Diego" and "imam" (a Muslim leader). And yes, the technology is developed to fight terrorists.

Existing search engines process individual documents based on the number of times a key word appears in a single document, Rohini Srihari, University of Buffalo professor of computer science and engineering explains. Instead, UIR is based on the construction of concept chain graphs that search for the best path connecting two concepts within a multitude of documents.

"A concept chain graph will show you what's common between two seemingly unconnected things," shes says.

“[The system] is designed to find the best path, the best chain of associations between two or more ideas. It returns to you an evidence trail that says ‘This is how these pieces are connected’”.

The research is supported by the Federal Aviation Administration and the National Science Foundation, and clearly the result of the post 9/11 increase in funding of anti-terrorist activities. Nevertheless, the goal is officially to implement the new technology in regular public search engines.

(This piece is partly based on an article by Lars Våge at Internetbrus.)

More search engine news...