CO-ODE HyOntUse
Collaborative Open Ontology User Oriented Hybrid Ontology
Development Environments
Development Environment
We want your feedback and ideas – Please join the community
web:
forum:
email:
http://www.co-ode.org/
http://www.co-ode.org/forum/
co-ode-admin@cs.man.ac.uk
Ontologies are formal methods for capturing knowledge about
domains of interest– biology, medicine, engineering, astronomy,
historical artifacts etc. They are based around representing
concepts (‘classes’) and the relationships between them
(‘properties’). Ontology development and management are critical
technologies for knowledge management and dissemination and
the foundation for the Semantic Web, and a new ontology
language, OWL, has just been approved as a standard by W3C.
Existing tools supporting ontology development and management
are primarily aimed at specialist "knowledge engineers".
The CO-ODE & HyOntUse projects aim to produce environments
for users to manage their own ontologies, plus plug-and-play
architectures for developers to build applications using them.
The teams behind OilEd (for OWL/DLs) and Protégé (for frames)
together with AKT and Epistemics are working with users to
produce a platform offering the best of breed for each community.
The new environment will exploit the existing Protégé
software architecture from Stanford University, providing
access to an established community of plug-in tool
builders:
• CO-ODE/HyOntUse are developing plugins to build on the
Protégé-OWL plugin from Stanford that will be:
–Intuitive enough to be used by domain experts with little
formal training.
–Flexible enough to support the specific needs of certain
user communities.
–Transparent enough to provide easily-used debugging
tools and explanations.
–Open enough to support several languages and
development paradigms (OWL, RDF(S), DL, Frames).
Our current prototypes provide ontology visualisation and support
common tasks through “wizards”.
The current prototypes and end user tools:
Frames and OWL/Description Logics (DLs) have much in
common, but they are more different than they look.
OWLViz
• Primitive concepts - Described but not defined and in a hierarchy
• Properties - relations between concepts and also in a hierarchy
• Descriptors - property-concept pairs
• Contingent Knowledge
- Defaults & Exceptions
• Reflective queries
• Individuals
• Hybrid reasoning
qualified by “some”, “only”, “at least”, “at most”
• Defined concepts
- Made from primitive concepts and
descriptors
• Axioms
- disjointness, further description
of defined concepts
• A Reasoner
- to organise it for you
OWL / DLs
• Meta data
OWLViz
provides the
user with a
view of the
ontology
before and
after
classification
for easy
comparison.
A first hybrid “logician’s environment” has been constructed at
Stanford to test the integration and requirements:
OWL Wizard
Asserted
hierarchy
Metadata
Axioms
Computed
hierarchy
Logical
definition
OWL Wizard
will provide
a plug and
play
architecture
for
supporting
task-based
actions on
the ontology.