Cognitive Adaptation of Management Behavior of Information via Observation (CAMBIO)

Increases in data volume and diversity place a disproportionate burden on decision makers, who must manually manage datasets with tools that are insensitive to their missions, tasks, and objectives. In time-critical situations, information management is more likely to consume decision makers’ time and attention than facilitate their analysis. In response, we propose to design and develop an adaptive workflow management system that flexibly responds to a user’s current tasking, information needs, and cognitive ability to interpret information. The system will learn, both actively (from user inputs) and passively (from user behaviors), what users need under different scenarios. Specifically, we will: • Provide a traceable suite of data preparation and exploration functions, enabling users to easily navigate their own workflows, and compose automated workflows for reuse. The system will learn from these traces how users invoke operations to manage information; • Build task-centric models of decision makers’ information needs, to predict immediate data management needs, and to respond to shifts that trigger changes in information management needs; and • Characterize cognitive state and state changes of decision makers, to automatically adapt levels of automation, system guidance, and complexity in information presentation to match the users’ specific information needs.