2005 ARDA Challenge Workshop


Title: Interoperable Knowledge Representation for Intelligence Support (IKRIS)
Technical Leader(s): Richard Fikes (Stanford University), Chris Welty (IBM Research)

Overview

The IKRIS Challenge Workshop was an 18 month project sponsored by the Disruptive Technology Office (DTO, formerly known as ARDA-Advanced Research and Development Activity). It started in April 2005, and concluded in September 2006. The IKRIS Challenge Workshop addressed the following challenge problems:

Enabling Interoperability. Knowledge representation and reasoning (KR&R) technologies have been developed under a variety DTO programs (including NIMD-Novel Intelligence from Massive Data) to provide advanced support services for intelligence analysts and decision makers. The development of new technology necessarily involves exploration of alternative approaches, and it is highly unlikely that any single KR&R technology will be developed that will be optimal for all purposes. However, systems that use differing (and potentially incompatible) KR&R technologies need to be able to interoperate in order to achieve DTO goals of providing effective broad-based information technology to support the intelligence community, and that interoperation will necessarily involve systems being able to interchange and use knowledge represented in other systems. Thus, there is a need to develop methods for achieving interoperability among systems that use multiple KR&R technologies rather than to dictate the use of a single KR&R technology.

Representing Intelligence Analysis Knowledge. KR&R applications use machine-processable encodings of knowledge to help intelligence analysts, e.g., sift through vast quantities of intelligence data, identify entities and relationships of interest, and formulate, test and refine their hypotheses. The knowledge in question may be about such things as external threats or situations, the analytic tasking(s) at hand, analytic processes, or the data being analyzed (e.g., its provenance, security classification, releasability, etc.). The KR&R challenges include the ability to represent a situation at a time point, a scenario produced by events and actions in situations, alternative hypothetical situations and scenarios, and incompatible claims made by different information sources. In addition, KR&R technology for intelligence tasks must address the core representation problems that arise because analysis tasks have open-ended domains (a task can be about most anything) and analysis necessarily involves considering atypical situations and events. All of this knowledge must be represented in an integrated manner that enables automated reasoning methods to access it and provide timely, reliable, and explainable results to an analyst, even when the knowledge is massive in multiple dimensions.

Workshop Goal(s)


Working Group Leaders


Workshop Participants

Bill Andersen (OntologyWorks); Craig Anken (AFRL); Fotis Barlos (BAE Systems); Danny Bobrow (PARC); Selmer Bringsjord (RPI); John Byrnes (FairIsaac); Brant Cheikes (MITRE); Steve Cook (USG); Andrew Cowell (PNNL); Chris Deaton (Cycorp); Keith Devlin (Stanford); John Donelan (USG); Michael Gruninger (U.Toronto); Ian Harrison (SRI); Robert Hoffman (IHMC); Jeff Hudack (AFRL); Mario Inchiosa (NuTech); David Israel (SRI); Charles Klein (Cycorp); Nancy Lawler (USG); Hua Li (Sarnoff); Arun Majumdar (VivoMind); David Martin (SRI); Mark Maybury (MITRE); Drew McDermott (Yale); Deborah McGuinness (Stanford); Sheila McIlraith (U.Toronto); Chris Menzel (Texas A&M); Dan Moldovan (LCC); David Morley (SRI); Leo Obrst (MITRE); Jennifer Ockerman (JHU/APL); Valeria de Paiva (PARC); Jean-Michel Pomarede (USG); Richard Rohwer (FairIsaac); David Silberberg (JHU/APL); John Sowa (VivoMind); Jennifer Turns (U.Washington); Marco Valtorta (U.South Carolina); Russ Vane (GD/AIS); John Walker (USG); Michael Witbrock (Cycorp); Wlodek Zadrozny (IBM)

Technical Products

NRRC Point of Contact: Dr. Brant A. Cheikes, bcheikes@mitre.org, 781-271-7505.

DTO Point of Contact: Dr. Rita M. Bush, rbush@casl.umd.edu.

 

Last Updated: August 2007