Supporting Net-Centric Operations through Bandwidth Friendly Intelligent Information Discovery and Filtering

We propose to address the growing challenge of information overload in network-centric information systems through the development of an agent based framework that improves the warfighter's awareness of critical information while simultaneously preserving those aspects that make systems like the Command Post of the Future (CPOF) so useful. In particular, we propose to enable effective information discovery in two ways: (1) enhancing multimedia data indices with context derived metadata to facilitate information rediscovery/reuse; and (2) exploiting simple models of warfighter tasks to drive a "fire and forget" and potentially proactive information discovery agent. We also propose to exploit similarly context-aware data filtering and integration techniques to improve the ability of S-3 and S-2 users to understand and exploit the available information. By exploiting the unique characteristics of battle command system data and task domain knowledge, our proposed system, InfoStalker, will provide a much better fit to the capabilities and requirements of the warfighter than general purpose information retrieval technologies. The InfoStalker concept was conceived in light of Stottler Henke's ongoing effort to analyze CPOF data archives. This foundation of knowledge together with a wide breadth of related reusable software components allows us to propose an aggressive Phase I effort.