Raytheon‘s BBN Technologies subsidiary is developing new tools for analyzing text and inferring meaning for a program sponsored by the Defense Advanced Research Projects Agency and the Air Force Research Laboratory.
BBN said its scientists are also working on tools for finding relationships and anomolies in the text Deep Exploration and Filter of Text program.
The subsidiary won $16 million in additional funding to continue working on the program.
“Much of the information that our analysts need is couched in ambiguous language,” said Prem Natarajan, BBN’s executive vice president of speech, language and multimedia. “The DEFT program aims to strip away that ambiguity and also to make it easier to integrate new facts into our existing knowledge bases in a meaningful way.”
Scientists are also conducting research and development work on tools for integrating new, individual facts into existing information stores.
BBN says analysts want to discover interconnections between disparate data elements and nuances in text.
The DEFT program also aims to develop capabilities for exploring and filtering natural language to find information and deliver alerts to users.
BBN is also working on a prototype system for DARPA that provides military commands translations and converts foreign text into English without analysts.