AFRL releases MEADE BAA
On July 18, the Air Force Research Laboratory (AFRL) posted a broad agency announcement for Multi-Source Exploitation Assistant for the Digital Enterprise (MEADE). For FY21 funding, AFRL recommends that white papers be submitted by March 27, 2020.
Seeking innovative research to develop a vetted digital assistant that provides analytical answers within operator-driven timelines. The software must perform constrained analytical tasks specified by user in the form of a question and include software agents that perform tasks independently (question answering).
The Air Force Research Laboratory, Information Fusion Branch, is soliciting white papers under this Broad Agency Announcement (BAA) for research, development, integration, test and evaluation of technologies/techniques to provide an interactive question answering system that functions as a virtual assistant by performing analytical tasks or services for an analyst.
Multi-INT analysis efficiencies need to be increased in order to match the complexity, velocity, variety, and volume of intelligence data being collected against increasingly agile and deceptive adversaries. The MEADE objective is to make complex analytics possible for nearly anyone, regardless of their technical ability. This effort is intended to not only support an Intelligence function, but also Command and Control functions (to make a decision). For purposes of MEADE, an ‘analytic’ is defined as a process that discovers, interprets and communicates meaningful patterns in order to describe, predict, or improve knowledge of Entities, Events, and Relationships in context.
The intent is to provide analytics that answer questions directly or that will interact with the user to help steer analysts to an answer rather than simply providing a ranked list of potential information sources to help the analyst answer the question themselves. In addition to directly providing information such as weather, calculations, or facts from existing sources or media, complex analytics shall be cued as needed. For purposes of MEADE, interactive question answering involves two aspects of “interaction”, one involves the mechanism by which you interact with the software (text, voice, file upload), the other relates to iteratively responding to user input as needed to reduce question ambiguity and improve product satisfaction. While fundamentally similar to chat-bots used in virtual assistants (Apple’s Siri, Google Assistant, Amazon Alexa, and Microsoft Cortana) as far as providing an artificial conversation, the actual intent is to use the conversation to improve the efficacy of analytical tasks important to the military against both polished Intelligence and raw sensor data. Ultimately, an airman would be armed with a richer set of information “connecting more dots” across air, space, cyber, land, sea, and undersea to provide a more comprehensive situation understanding.
MEADE involves creating the foundation for a fully functional “Information Wingman” by adding the element of mutual support to enhance analytics. The intended deliverables include software tools, interfaces, and processes that are used by an analyst to direct, clarify, and scope automated analytics.
This will be achieved by pursuing two Focus Areas (FAs), Real-Time Operator-Driven Gist Exploration and Response (ROGER) in conjunction with Interactive Analytics and Contextual Fusion (IACF).
Full information is available here.