Fighting terror with knowledge
Fighting terror with knowledge
The US Central Intelligence Agency (CIA) is in the process of developing a "document exploitation' system that will improve its ability to analyse key information needed in the battle against terrorism. By Rodney Appleyard.
Despite being under fire recently for its failure to provide accurate information about the 9/11 attacks and weapons of mass destruction, the CIA hopes that its latest information management program will make a huge difference to its ability to analyse masses of information quickly.
The CIA has been overwhelmed by a tidal wave of information from many different sources in recent years, and this has escalated significantly since the War on Terror began. The agency has also experienced information overload when dealing with the hunt for weapons of mass destruction and keeping track of prisoner records.
Large volumes of information have placed huge demands on the IT infrastructure and intelligence staff and the international scope of these research activities has also placed a strain on a limited pool of experts. Getting the correct information to the right people quickly has been extremely difficult and costly.The CIA plans to use Kofax Mohomine and Ascent Technologies to help it cope with these problems.
The combination of these two solutions will allow valuable, specific information to be extracted much more quickly from masses of documents so that action can be taken much more effectively by the relevant post of command.
Ascent enables companies to collect large volumes of forms and documents, transform them into useful, retrievable electronic information and deliver it all into a variety of workflow document management systems. In addition, it can also capture e-documents and XML streams.
The Mohomine technology, developed by Kofax, which is part of the DICOM group of companies, offers automated text categorisation and extraction capabilities that enable the intelligence community to convert these vast amounts of printed and electronic documents, captured through Ascent, into sharable and exploitable information, which could essentially save lives if it allows the CIA to infiltrate terrorist activities.
The deal has been made directly with In-Q-Tel on behalf of the CIA. In-Q-Tel is a private, independent enterprise funded by the CIA. Launched in 1999, In-Q-Tel's mission is to identify and invest in companies that develop cutting-edge information technologies to serve the United States' national security interests.
Gilman Louie, In-Q-Tel's chief executive officer, says that the Kofax agreement will be successful in its mission of turning large volumes of disorganised paper documents into readable, prioritised information.
"Finding the critical pieces of information in the tons of hard copy data gathered today is an important challenge for the intelligence community.
Currently, analysts must manually sort through very large volumes of unorganised documents to find the valuable information that may offer the missing piece of an intelligence puzzle. We expect that Kofax' document exploitation technology will help solve this problem by enabling organisations to quickly capture all of their unstructured paper documents and automatically sort, prioritise and route content.'
The CIA has released statistics which illustrate the extent of this information problem that it has been plagued by recently. Under "Operation Just Cause', initiated in December, 1989, six million documents were created, weighing 100 tons and covering half a mile in stack. 12 million documents were created under Operation Desert Storm from January, 1991, weighing 200 tons and covering 1.1 miles of stack. Operation Iraqi Freedom, started in March 2003. It includes 78 million documents (so far), weighing 1,300 tons and covering a 7.0 mile stack.The aim of this new solution is to collect and process these complex paper documents, transform them into categorised, translated & actionable intelligence and then deliver this intelligence to the appropriate resource.
The highest priority documents will cover weapons of mass destruction, troop movements and prisoner records. The lower priority materials will include weather reports, city planning and transportation.
It is planned that large volumes of these paper documents will also be turned into mission-critical information for fast, efficient analysis and action.
The new system will automatically prioritise documents based upon textual content. The text categorisation solution uses pattern recognition technology to sort documents into user defined groupings and the technology has the ability to read and sort critical languages, such as Arabic, Pashto, Farsi, Dari and Chinese, into categories if necessary.
Congresswoman Susan David, who represents the San Diego-area 53rd Congressional District, says this technology is crucial to the war on terror. She is also a member of the House Armed Services Subcommittee on Terrorism, Unconventional Threats and Capabilities, which is responsible for the Department Of Defence counter proliferation and counter terrorism programs and initiatives. The subcommittee directs DOD information technology policy and programs, force protection policy and oversight, and related intelligence support.
"With our country's ongoing battle against terrorism, we must make it easier for analysts to comb through the large volumes of unorganised documents to find a missing piece of an intelligence puzzle. The Mohomine technology is needed to help solve this problem by enabling intelligence agencies to quickly capture all of their unstructured text documents, then automatically sort, prioritise and share the information. It is with the greatest pleasure that I have watched Mohomine mature from an emerging technology in my hometown of San Diego to become such a vital contributor to our nation's security.'
Sameer Samat, the Chief Technology Officer at Kofax discusses how too much information has caused problems for the CIA. "Today, intelligence organisations must cope with enormous amounts of unstructured data that comes from a wide array of sources, such as Web pages, news articles, email, cables, and innumerable printed and handwritten document types. It can be overwhelming to manually sort through these documents to glean specific data needed for intelligence-analysis purposes, and that is where text categorisation and extraction technologies such as Mohomine can be invaluable resources.
"The information is collected from a number of different sources and then programmed into the system. Each intelligence organisation can customise the information to make it easier for them to retrieve vital information. The system saves on time and prevents information from getting congested. Scanned images are also cleaned up. The processing speed is incredible. It can digest a 3000 page novel in a second.'
The CIA expects the system to eliminate the labour demands of having to manually index documents so that staff can spend this time on higher prioritised activities instead. It hopes that formerly inaccessible information will now be more available to drive intelligence and analytics so that it will be able to track down terrorists more quickly because the analysis of information will be carried out much more thoroughly.
Related Article: