a comprehensive framework for building multilingual domain ontologies
Upload: aims-agricultural-information-management-standards-fao-of-the-un
Post on 15-Jan-2015
426 views
DESCRIPTION
TRANSCRIPT
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 03
A Comprehensive Framework for Building Multilingual Domain Ontologies:
Creating an ontology on Food Safety, Animal and Plant Health
(OFsAPH)
Boris Lauser
Nordic AOS Workshop: Copenhagen 28th February 2003
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Agenda• Introduction:
– Why: The Biosecurity Portal Project– How: The modeling approach
• Framework for ontology creation• Application of framework:
– Creation of the Food Safety Ontology • Outlook:
– Application scenario
• Discussion
Introduction
Framework
Application
Outlook
Discussion
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
The IP-FsAPH International Portal on Food Safety, Animal and Plant Health
• access point for official national and international information on biosecurity
• interdisciplinary approach
• integrated access to information in the 3 areas
• global public access
• controlled access to nationally nominated users
Currently available on: http://193.43.36.96/servlet/CDSServlet
Introduction
Framework
Application
Outlook
Discussion
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
The IP-FsAPH International Portal on Food Safety, Animal and Plant Health
Introduction
Framework
Application
Outlook
Discussion
• Provides access to large amounts of data, coming from various resources from all over the world
• Need to make this data available and searchable through the portal
• Realization by exposing metadata
• Need for controlled, commonly agreed on subject vocabulary
• Integration of an ontology to provide the necessary controlled vocabulary and semantics which can be explored for enhanced information retrieval
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Introduction
Framework
Application
Outlook
Discussion
KAON The Karlsruhe Ontology and Semantic Web Tool Suite
KAON is an open-source ontology management infrastructure
Major Components:
• OIModeler: tool for ontology creation and evolution• KAON Portal: a web based portal for browsing KAON ontologies• KAON API: a programming API to access the ontology
independently from any storing mechanism• Engineering Server:
ontology storage mechanism based on relationaldatabases to provide concurrent access and scalability
• Text-To-Onto: Semiautomatic ontology creation using text miningtechniques
Freely available on: http://kaon.semanticweb.org
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Introduction
Framework
Application
Outlook
Discussion
The Generic RDFS model:
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
The KAON modeling approachIntroduction
Framework
Application
Outlook
Discussion
The KAON lexical model extension:
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Introduction
Framework
Application
Outlook
Discussion
Agenda• Introduction:
– Why: The Biosecurity Portal Project– How: The modeling approach
• Framework for ontology creation• Application of framework:
– Creation of the Food Safety Ontology
• Outlook:– Application scenario
• Discussion
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
The framework
A comprehensive frameworkfor building a domain ontology
Focus:Concept acquisition and developmentof the lifecycle of ontology creation
Introduction
Framework
Application
Outlook
Discussion
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
The framework: 5 phases
• Resource selection• Semiautomatic ontology concept
acquisition– Creation of a core ontology from scratch– Reuse of existing vocabularies
• Merging of ontologies• Extension and refinement• Evaluation
Introduction
Framework
Application
Outlook
Discussion
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Coreontology
Manualcreation
FocusedWeb crawling
List of domainstart web
pages
List offrequent
terms
List of domainSpecific
documents
Term BT t1 NT t2 RT t3Term USE t3…
Thesaurus
RDFS ontologymodel
convert Ontology pruning and learning
algorithm
Domaincorpus
Genericcorpus
Prunedontology
List of critical
concepts
Semi-automaticOntology
Acquisition
Mergingof
ontologies
Refinementand
Extension
Evaluation
Selectionof resources
Manual creationof core ontology
1st acquisitionapproach
2nd acquisitionapproach
Text To Onto
The Framework: overview
Introduction
Framework
Application
Outlook
Discussion
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Introduction
Framework
Application
Outlook
Discussion
Agenda• Introduction:
– Why: The Biosecurity Portal Project– How: The modeling approach
• Framework for ontology creation• Application of framework:
– Creation of the Biosecurity Ontology
• Outlook:– Application scenario
• Discussion
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Semi-automaticOntology
Acquisition
Mergingof
ontologies
Refinementand
Extension
Evaluation
Selectionof resources
Manual creationof core ontology
Application of the framework:1st iteration
Introduction
Framework
Application
Outlook
Discussion
1st iteration
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Phase 1: Selection of Resources/ Manual creation of core ontology
67 concepts91 relationships
Information Resources:•Brainstorming•Codex Alimentarius•SPS Agreement
Core Ontology
Ontology Editor(OIModeler)
3 subject specialists
Introduction
Framework
Application
Outlook
Discussion
1st iteration
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Phase 2: 1st Acquisition Approach:
Focused Crawling
Focused Web Crawling
68 concepts91 relationships
Core Ontology
List of extracted main sites:http://www.foodsafety.gov/ Gateway to Government Food Safety Information
http://vm.cfsan.fda.gov/ Center for Food Safety & Applied Nutrition
http://www.inspection.gc.ca/ Canadian Food Inspection Agency
http://www.extension.iastate.edu/foodsafety/ Iowa State University - Food Safety Project
http://www.foodsafety.iastate.edu Iowa State University - Food Safety Consortium
http://www.fsis.usda.gov/ United States Department of Agriculture, Food Safety and Inspection Service
http://www.nal.usda.gov/foodborne/index.html Foodborne Ilness Education Information Center
http://www.euro.who.int/foodsafety World Health Organization – Regional Office for Europe Food Safety Programme
List of 257 food Safety domainweb pages
Grouping into Main sites
Introduction
Framework
Application
Outlook
Discussion
1st iteration
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Phase 2: 2nd Acquisition Approach:
Thesaurus Pruning
Food SafetyDocuments
GenericDocuments
Rice BT … NT … RT … RT … RT … …
AGROVOC27365 keywords
Automatic Pruning
Extracted ontological structure:# of concepts: 504taxonomic depth: 5
5 evaluation runs
1632 frequent terms
Introduction
Framework
Application
Outlook
Discussion
1st iteration
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Phase 3/4: Merging of Ontologies, Refinement
1632 Terms from pruning process 12 new concepts
extracted
Ontologicalstructureextracted from AGROVOC
23 new conceptsWith hierarchicalrelationships extracted
67 concepts91 relationships
Core Ontology
Assemblystep
92 new relationshipscreated
Biosecurity OntologyPrototype
102 concepts183 relationships
Introduction
Framework
Application
Outlook
Discussion
1st iteration
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
• Open to users and subject specialists for evaluation
• http://localhost:8080/faoportal/dispatcher
Introduction
Framework
Application
Outlook
Discussion
1st iteration
Biosecurity Ontology Browser Modified version of the KAON Portal
Phase 5: Evaluation
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Semi-automaticOntology
Acquisition
Mergingof
ontologies
Refinementand
Extension
Evaluation
Selectionof resources
Application of the framework:2nd iteration
Introduction
Framework
Application
Outlook
Discussion
2nd iteration
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Phase 1/2/3 : Resource selection, acquisition, merging
Biosecurity OntologyPrototype
102 concepts183 relationships
Text To Onto ~100 domain
Specificdocuments
AGROVOC
Revised OntologyPruner
List offrequent
terms
Pruned Agrovoc: ~3000 concepts
Ontology Editor(OIModeler)
Merging &Refinement
1st acquisitionapproach
2nd acquisitionapproach
2nd iterationIntroduction
Framework
Application
Outlook
Discussion
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Phase 4 : Extension and Refinement
BiosecurityOntology
Core
2nd iterationIntroduction
Framework
Application
Outlook
DiscussionGeographic Area
Ontology
Generic PropertiesOntology
Ontology on Food Safety,
Animal and Plant Health
• 3761 concepts• 16 unique relationships
IPPC glossary
• creation of a modular design for reusability
Further Codex Alimentarius
Classifications
OIE classifications
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
AgendaIntroduction
Framework
Application
Outlook
Discussion
• Introduction:– Why: The Biosecurity Portal Project– How: The modeling approach
• Framework for ontology creation• Application of framework:
– Creation of the Biosecurity Ontology
• Outlook:– Application scenario
• Discussion
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Application scenario: 2 use cases
Use Case 1: Indexing the subject of a document
Use Case 2: Searching information on the portal
Risk;…Subject
Title
…
…
OFsAPH
Risk;…Search…
…
Indexer
Searcher
Introduction
Framework
Application
Outlook
Discussion
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Search:
Risk assessment
Biosecurity Portal:
…
…
OntologyEnabled Search
Application
Display
Ontology Metadata + Doc base
Introduction
Framework
Application
Outlook
DiscussionSearch Results
+Ontology SemanticsKAON API
Simple query
Ontologysemantics Enhanced
query
Found results
Use Case 2: Ontology based search extension
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Introduction
Framework
Application
Outlook
Discussion
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Introduction
Framework
Application
Outlook
Discussion
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
Introduction
Framework
Application
Outlook
Discussion
FAO of the UN
Library and Documentation
Systems Division
Nordic AOS Workshop
Copenhagen
February 2003
AgendaIntroduction
Framework
Application
Outlook
Discussion
• Introduction:– Why: The Biosecurity Portal Project– How: The modeling approach
• Framework for ontology creation• Application of framework:
– Creation of the Biosecurity Ontology
• Outlook:– Application scenario
• Discussion