Transcript Intelligence

Semantic Web for the Military User
Transition Breakout Session—Intelligence
Dr. Joe Rockmore/Cyladian Technology Consulting
Participants
•
•
•
•
•
•
•
•
•
Pamela Arya/NRO
Tim Finan/UMBC
Lee Lacy/DRC
Richard Lee/Logicon
Tom Gower/ONI
David Martin-McCormick/IMO
Bob Neches/USC-ISI
Joe Rockmore/Cyladian Technology Consulting
Al Schuler/Aerospace
Value Propositions
•
•
•
•
Consumers = custom products
Producers = get credit for production
“Query mining”
Feedback from missing information, including to
collection management
• Feedback on use of marked up data
– Hit counts are poor, but easily measured
– Can measure demand
• Consolidation of data
• Publication once, derived products
Significant Issue: Geolocation &
Temporal Representation
• Understand document enough to know locations
mentioned in a document
• Placename, lat/lon, BE num, UTM, etc. disambiguation
Significant Issue: Markup Tools
• Consumer-based markup tool needed soon
• Culture = analysts too busy to do any more work,
including markup, unless
– Its very easy to do
– There is clear value to producers (not consumers)
– Someone measures them on the quality/quantity of markup
• Produce knowledge objects from the outset, format from
these objects, including English text documents
– Will only work in limited cases, when reports are sufficiently
structured
– Expressibility limitations at odds with identifying the unusual,
which is an important task in intelligence
– Make key points as knowledge objects, embellish with natural
language & use embellishment to improve ontologies
Significant Issue: Access to Data
• Tailored push; also pull (“My Intelink”), including changes
of sufficient magnitude
– Subscriptions and data descriptions for matching against
subscriptions may be best done using hierarchical ontologies
•
•
•
•
•
Crawlers of value, but may have access control issues
Uncertainty of data (both by source and about source)
Inference-based retrieval of information
Pedigree critical to maintain
Indexing of markup important for speed of access
Significant Issue: Collection
• Tie collection, processing, production together
• A common markup language will enhance collection, thus
optimizing use of intel sources
• Producers and consumers have different ways of looking at
the world, not necessarily a mapping between them
– Can consumers provide tasking to producers, via markup of
requirements on collection?
– Info data needs from UJTL tasks or other statement of data needs
• Will DAML markup allow semantic understanding of
information enough to affect releasability processes?
Recommendations
• Military and intelligence users, beyond those at this
workshop, that particularly should hear about DAML:
–
–
–
–
–
NIMA Agent-Based Initiative
Information consumers (e.g., service ops centers)
SOCOM
SPACECOM, NIPC (computer network defense)
NCS
• Recommendations—How do organizations understand
what DAML products/approaches could help them?
– Focused TIE’s with appropriate producers and consumers around
specific value propositions
– Organize DAML web site by functionality; describe capability,
maturity, etc. Make more useful. Also, need contract approach
synopses and status.