Anti-Terrorism Search Technology Helps Uncover Covert Content Threats

Article Tools

• Bookmark

Advertisement

Most Popular News

• Accurately measuring pay in the public sector
• Government spending legislation aims to audit GSA each year
• Investigations could lead to tighter controls on agency spending

Know before you buy or sell. Get your truck's value instantly online.

Find Your Truck Now

*To see the actual values, you must purchase a subscription or a single-vehicle lookup.

IBM, Armonk, NY, announces a new search technology that enables intelligence and law enforcement agencies to uncover hidden patterns to identify potential criminal or terrorist activity. The solution can find covert information by analyzing text within surveillance transcripts, Web pages, blogs, Internet news feeds,
e-mail, and other “unstructured” content sources.

The solution is based on IBM’s new open framework for text analytics called Unstructured Information Management Architecture (UIMA), which provides the platform for processing, indexing, and searching information for enhanced understanding. UIMA is the result of more than four years of development by IBM Research and Defense Advanced Research Projects Agency (DARPA), the central research and development organization for the Department of Defense.

Until now, search solutions could only find covert documents based on key words in the documents themselves. They typically could not find hidden meaning and relationships within text. Using this new solution, a search for “George Bush in Israel,”for example finds documents about George Bush visiting Israel, not just documents with “George Bush” and “Israel” in the text.
Also, most search terms are incomplete and ambiguous and can’t make inferences about the true intent of users’ search terms. This new solution can distinguish between different semantics of the same term [e.g. rock (stone) vs. rock (music) vs. rock (to move back and forth)].

Local and federal agencies can use the solution to analyze a wide variety of information sources, such as field reports, ship manifests, and Internet-based information, along with content delivered by Factiva, including press releases, news articles, and financial data from more than 9,000 business news and information sources. Content is analyzed through text analytics software from Attensity to identify people, organizations, actions, locations, events, and related associations.

Download the UIMA framework at www.govinfo.bz/5195-159.