Raytheon's BBN Technologies subsidiary and Charles River Analytics have partnered to create tools that will work to help military analysts classify bias at the sentence and speaker levels.
Charles River Analytics said Thursday it will apply a linguistic feature-based machine learning approach and a grammar toolkit in efforts to develop bias classification technology under the Defense Advanced Research Projects Agency's Causal Exploration program.
DARPA seeks a causal modeling platform for military planners to determine what factors drive a conflict situation.
Robert Hyland, principal scientist at Charles River Analytics, said the partnership with Raytheon BBN seeks to establish the foundation for a system to detect propaganda and information activity.
“The U.S. military needs better methods to deal with adversary scale and complexity; these adversarial techniques interweave into open and free press and information spaces,†Hyland added.
Charles River Analytics used natural language processing techniques to trace, analyze and address information signals from technical documents and news briefings under the THREAT and ALARMM projects.