Transcript One Page PowerPoint “Quads” on this White Paper

BAA Number: BAA 04 04
White Paper Control Number: SGIB
Title: Architecture for Managing Incident Information (AMII)
Data mining of
unstructured text:
One of our preliminary
deployments as a
taxonomy generation tool
for the FCC
Nine Step Process Model:
Creates structural ontology
through a real time
measurement of information
flow and uses this to faceplate
the communication of all
essential information
Proposed Technical Approach:
The problem is informational transparency within a secure environment.
This problem is approached using a optimal data encoding format which
has no proprietary limitations. Several existing knowledge management
or information creation software products will be modified to interact with
a core environment operating system based on a minimal Linux kernel
and the a Ontology referential base (Orb). Telecommunications device
interfaces to PocketPC and phones will be developed. Visualization
software is to be integrated and user training materials developed.
The approach also has a non technical human-factors aspect. Deep and
broad curriculum will be developed over the concepts of knowledge
creation and sharing within communities of practice. This will not be a
marketing tool, but will be university curriculum participated in by faculty
from several universities and form several disciplines.
Principle Investigator:
Peter R. Stephenson
25 October 2004
Operational Capability:
An underlying simplification of a data encoding standard has been
developing over the past decade and is being demonstrated in our
prototype, and in systems that we have knowledge of. This simplification is
provably optimal in terms of encoding and data mining over relational data.
A seamless integration occurs when everyone uses this optimal solution.
Information assurance (IA) follows when secure channels are designed for
the flow of information having this form, and the management of
information becomes unencumbered by technical problems related to
software incompatibilities. Legacy data can be easily harvested. Human
language can be associated with structured ontologies and the means to
develop these resources provided to the user. Representation of the global
and specific conversation are encoded into a visual form using graph
theory. Geographic and equipment data can be encoded in the same form.
Information sharing develops out of need and in context, and is not
inhibited by information technology infrastructure limitations.
The capability is in the advanced prototype stage. Deployment and
transition requires the development of a curriculum and training experience
in the incident management arena.
Cost and Schedule:
§ Quarter 1: A series of workshops will be held and advanced benchmarking and feasibility
tests made. Collaboration with State of Michigan departments responsible for the human factors
interface to first responders. Quarter 2: Technologies will be tailored to incident command
communities and first responder communities. Quarter 3: A series of workshops will be held and
prototype implementations of the proposed technologies will be demonstrated and critiqued.
Real world simulations will be offered and training requirements studied. Quarter 4: The
prototype implementations will be updated and simulated scenario deployments demonstrated to
HSARPA. A work plan for implementation and deployment of a production version of the system
will be prepared and submitted for approval. First year cost: $946,000
Deliverables:
Several knowledge management or data mining software systems will be integrated based on
the common data encoding, called Orbs. The Prototype will develop a small distributed
operating environment consisting of computers, phones and PocketPC. Training materials and
educational curriculum will be developed. Specific scenario based situational ontology will be
developed for re-use. Reports will be generated. Field training, test deployments and transition
programs will be delivered.
Contact Information:
Principle Investigator: Peter R. Stephenson
The Center for Regional and National Security, Eastern Michigan University
[email protected]