Rand offers architecture for automating data analysis
Connecting state and local government leaders
Rand Corp.'s Atypical Signal Analysis and Processing Architecture could help intelligence agencies better spot potential terrorist activity.
Rand Corp. of Santa Monica, Calif., has developed an architecture that intelligence agencies could use to better spot potential terrorist activity, according to John Hollywood, an operations researcher for the nonprofit organization who helped craft the architecture.
Other agencies could also use the architecture in their domains, Hollywood said.
'We specifically designed it for counterterrorism, but a lot of the overall approach would apply to a wide range of intelligence analysis,' he said.
The Atypical Signal Analysis and Processing schema is 'an overall architecture for taking streams of information and going through a number of analysis steps, in a highly iterative, context-dependent way that eventually flags significant results,' Hollywood said.
The architecture sorts data through several steps. Data about suspicious people, places, things and financial activities is first collected from a number of government external databases. This information would be supplemented by field reports of unusual activity.
Software agents then scan this information looking for relationships and then form hypotheses about how the different bits are related. The system then prioritizes its findings and alerts human analysts of the most important cases.
'You're filtering incoming information, flagging the most useful pieces, linking incoming data elements together and generating hypotheses of the information,' Hollywood said.
Commercial software already offers a large amount of this functionality. What ASAP offers is a way to fit the pieces of software together into an overall system that automatically takes the first steps of analyzing data.
An agency could implement such an architecture over time, Hollywood said. First, the agency might standardize how analysts record data, then set up posting boards to share information. Moderators could review the boards and cull the interesting bits. Over time, the agency could start feeding information into this review process from automated sources.
This is not the first attempt at designing an automated analysis tool for government intelligence gathering. The Defense Advanced Research Projects Agency's Total Information Awareness concept had a similar scope, but was scrapped due to privacy concerns.
ASAP would be different from TIA in that it would not rely on information from private industry'such as credit cards records and travel records'but rather on information that the government has already collected. It also would differ in sheer size from TIA; whereas TIA comprised multiple sources, ASAP would only tap into databases, not duplicate their contents.
'This is not a massive intelligence storage exercise. If the ASAP database that you are working from does not fit on a single laptop, then there is a problem,' Hollywood said.
Rand funded the project in-house and plans to present the architecture to government officials in the upcoming weeks.
NEXT STORY: DHS to update its collaboration portals