Ontology Terminology

1
OntologyTerminology

• Sergio DElia
• Head of Services Infrastructure Technology
• ESA - EOP Ground Segment Department

2
Objective / Approach

• ObjectiveWiden the use of the EO products,
  applications, services
• Approach (one)Permit easy, semantic
  identificationfrom non-EO domains (using domain
  specific terms)of relevant EO services,
  applications, data or information

3
Suggested Method

• Specify at high level the EO reality EO Ontology
• Add classification vs. other domains
  Multi-domain Ontology
• Derive a simplified, more abstract Classification
  Ontology
• Supporting multiple domains
• Providing equal visibility of all concepts
• Using fixed concepts and relations
• Permitting an implementation
• as an isolate Web Service
• with limited dependency from evolving reality /
  dynamic changes
• Verify its applicability (also to non-EO domains)

4
EO Ontology
Concepts
Relations
5
Multi-domain Ontology
6
Simplify / Abstract

• Apply definitions and replace real concepts with
  abstractions (categories)
• Service Service Category
• Application / Processing Application Category
• Data / Information Product Category
• Local structures will include the other concepts
  / links(Product, Application, Service)
• In local terms (for correctness / compatibility
  verification)
• With explicit links to Product, Application or
  Service Categories

7
Classification Ontology
8
Verification

• EO domain classification ontology applies to any
  domain
• The deepest concept (Product Category) can be
  reached from any domain
• The concepts are fixed
• A stable, Web Service implementation is possible
• The TN Ontology and Terminology for EO and other
  Domains provides preliminary verification for a
  few cases

9
Knowledgebase

• Instantiation of the Classification Ontology
• To contain instances of
• Concepts
• Links (weighted for navigation and confidence
  level)
• Instances will evolve with use
• Tools needed to trace user paths for Knowledge
  Discovery

10
Multi-language Support Proposal

• Knowledgebase in English
• One layer for each other language(exactly
  replicating English tables, with one to one
  translations)
• Searches / interactive navigation
• Translate terms to English
• Search knowledgebase / retrieve results
• Translate retrieved terms to user language
• Links (stored only against English) represent
  average values across all uses (mirror of the
  international community)

11
Example Tables