INTENT OF WORK
FOR
COMPONENTS FOR ONTOLOGY DRIVEN INFORMATION PUSH (CODIP)
Contract # F30602-00-C-0192
Prepared by
GRC International, Inc
.,An AT&T Company
19 March 2001
TECHNICAL POC (PRINCIPAL INVESTIGATOR):
Lewis Hart
GRCI, an AT&T Company
1900 Gallows Road
Vienna, VA 22182
703-506-5938
CODIP's Goal
The Components for Ontology Driven Information Push (CODIP) program, as the name implies, is focussed on the distribution of information based on explicit semantic content. A key goal of CODIP is to develop technology and tools to build and use articulation ontologies. An articulation provides information to map terms and concepts from one ontology into a second ontology. The articulation can be used, for example, as the basis for searching across multiple domains, each with their own separate ontology. It can also be use for translating message facts from one ontology into a recipient's desired ontology. Finally, articulations enable the integration of multiple-source operational database systems into an integrated data warehouse/data mining system.
Products and their Users
Three primary product areas support these goals and are under development:
Tools
CODIP tools are focussed on supporting application builders applying DAML technology to their systems. The three principle tools being developed are:
DAML UML Enhanced Tool (DUET )
DUET provides a UML visualization and authoring environment for DAML. Core DAML concepts are being mapped into UML through a UML profile for DAML. DUET will have the capability to work with multiple ontologies simultaneously, and to interact with the Ontology Articulation Builder (OAB) services to interactively build articulations between ontologies. Its intended users are database designers and systems engineers, many of whom already have a good understanding of UML and object-oriented modeling, which can be leveraged to apply DAML to their systems.
Ontology Fact Processor (OFP)
The OFP will provide a publish and subscribe service that uses DAML articulations and fact translations to provide specific content routed to consumers in their specified ontology. This capability can be used by system developers to automate database translation and integration, provide explicit semantics for web-izing of legacy databases and bridging applications between domains of discourse.
Hot DAML Project
Our Hot DAML project will leverage IR&D funding and existing GRCI developed GOTS applications to build a distributed, web centric RDBMS meta-data analysis and management tool. The target users of this system are data stewards, and data integration system developers. The purpose is to add DAML import/export and articulation generation to existing database analysis and integration tools which have an existing customer base, as well as to provide a web-based user interface for data mapping.
Services
CODIP services will provide capabilities that support users, applications and agents navigating and searching the web. Three of the DAML-centric services we are developing are:
Ontology Articulation
This service will provide automated analysis of potential mappings between ontologies and build articulation ontologies that codify the mappings in DAML.
Fact Translation
Fact Translation services will provide translation of statements from one ontology into equivalent statements in another ontology based on articulation ontologies.
Web Directory
The Web Directory will provide an integrated access point to directories of web resources. It will be built using the Ontology Articulation and Fact Translation services.
Components
GRCI is developing a suite of DAML components that provide the foundation for the above tools and services. The primary users of these components will be the DAML application developer community.
DAML API
The DAML API provides Java interfaces and utility classes for manipulation of DAML ontologies.
UMM API
The UML Meta-Model (UMM) API provides access to DAML as mapped into UML. The UMM is designed to facilitate the UML visualization and analysis with existing third party UML Tools.
Articulation API
A third interface has been identified and is being designed. It will extend the DAML API to include specific analysis and manipulation capabilities, including fact and ontology translation.
MARIA Components
MARIA is a distributed agent architecture that provides a flexible distributed architecture for building component-based software agents. Several components will be developed and will be integrated with the MARIA environment, however they will also be usable outside of MARIA.
Ontology Articulation Builder (OAB)
The OAB will provide an analysis of similarities between ontologies. The analysis utilizes explicit information (thesauruses, other ontologies), implicit information (structure, data-types, known patterns) and human guidance to produce articulations ontologies.
Information Push Engine
The Ontology Fact Processor is built around an information push engine, implemented as a collection of MARIA Behaviors that provide a semantics-based information push system.
DAML Rule Engine
The Java Expert System Shell (JESS) developed at Sandia Laboratories has been integrated in to the MARIA environment. When used with the DAML API, it provides a rule-based reasoning environment for DAML.
Transition and Collaboration
Transition
In addition to the HORUS and CALL transition targets, GRCI has identified several, specific additional opportunities to apply DAML technology. GRCI is actively pursuing the application of DAML technology to the following programs*:
GRCI is also working with several other organizations which are likely to have applications for DAML, including DISA/DSO for data standardization, data segmentation in the DII/COE, and the Defense Model and Simulation Office (DMSO) for XML issues in its High Level Architecture.
Collaborations
We have identified several birds of a feather (BOF) groups and teams for collaboration.
Graphical BOF
Working with Lockheed Martin UBOT program to define UML representation of DAML, and to influence the UML 2.0 specification to be more supportive of ontology modeling.
Lifecycle BOF
Working with the lifecycle BOF group to deploy CODIP tools--DUET for visualization and analysis, and our Hot DAML project for distributed meta-data/ontology analysis and management.
Services BOF
Working with the services BOF group (Cycorp and Stanford KSL) to define the representation of services and to provide DAML services on the web.
Translation BOF
Working with the Stanford Database Group to apply their automatic articulation development research to automatic articulation development in our OAB and with other translation BOF members (Tecknowledge) for ontology translation.
Future Vision
In the short term, GRCI will be developing and will begin releasing to other DAML researchers the products discussed above. GRCI will also be actively pursuing the transition of DAML products to funded early adopter projects, such as ATD and ACT II programs. In the longer term, GRCI will deploy DAML technology in military information systems. Several opportunities have been identified that will be awarded in the next nine to eighteen months, for example GTN21 and SDE. Furthermore, our Tools and Services will be available on the WWW for use by our own and third party DAML aware applications. The tentative milestones for the CODIP product releases are shown in the table.
Tentative Milestones |
||||
Product Areas |
Next 3 Months |
3 to 6 Months |
6 TO 12 Months |
Beyond 12 months |
Tools |
||||
DAML UML Enhanced Tool (DUET) |
Rel 1 |
Rel 2 |
||
Ontology Fact Processor (OFP) |
Rel 1 |
Rel 2 |
||
Hot DAML Project |
Rel1 |
Rel 2 |
||
Services |
||||
Ontology Articulation |
Rel 1 |
|||
Fact Translation |
Rel 1 |
Rel 2 |
||
Web Directory |
Rel 1 |
Rel 2 |
||
Components |
||||
DAML API |
Rel 1 |
Rel 2 |
||
UMM API |
Rel 1 |
Rel 2 |
||
Articulation API |
Rel 1 |
Rel 2 |
||
Ontology Articulation Builder (OAB) |
Rel 1 |
|||
Information Push Engine |
Rel 1 |
|||
DAML Rule Engine |
Rel 1 |