350 likes | 364 Views
Learn about FAO's projects on Semantic Technologies and the AGROVOC/CS Workbench. Explore terminological resources, ontologies, and future works in knowledge exchange and capacity building.
E N D
FAO projects in the area of theSemantic Technologies 23rd APAN Meeting Manila, Philippines margherita.sini@fao.org
Outline • Introduction • The AGROVOC/CS Workbench • Future works
FAO as a Knowledge Organization • Knowledge Exchange & Capacity Building Division • Teminological resources (AGROVOC, FAOTERM, Glossaries, etc.) • Avian Influenza Glossary (Avian Influenza/Pandemic Influenza Glossary in collaboration with World Health Organization) • Subdomain ontologies • Agricultural IM standards: http://www.fao.org/aims
Background (1/2) • AGROVOC • Used worldwide • Multilingual (online 11 languages) • Term-based (as opposed to concept based) • Limited semantics (NT, BT, RT, USE/UF) • Maintained as a relational database • Distributed in several formats (RDBMS, TagText, ISO2709, ...)
Background (2/2) • Draft versions available in TBX, SKOS, OWL • Access to full thesaurus through seven Web Services • Agricultural Ontology Service (AOS)
AGROVOC/CS Workbench: a collaborative ontology development tool • Support and manage the multi-language terminology work of information management specialists in the development, maintenance, and quality assurance of the CS.
AGROVOC/CS Workbench: Users • Term editors • Ontology editors • Validators • Publishers • Administrators distributed user-base!
Text processing Corpus Creation Corpus Analysis Manage Concepts, Terms, Relationships Classification Schemes Quality Assurance Other functionalities Search Import / Export Validations Administration Help AGROVOC/CS Workbench: Features
Concept Hierarchy text corpus .doc, .pdf, .xml, etc. AGROVOC/CS Workbench: Overall design input AOS/CS Workbench concordance pattern-matching multilingual input
Current status • AGROVOC/CS Workbench construction (ongoing) • CS OWL model described • Test project available • Full AGROVOC conversion procedure using the OWL model (ongoing) • Relationships definition (in collaboration with CNR, Italy) (ongoing) • Performance tests (ongoing)
Next steps • Completion of the AGROVOC/CS Workbench • AGROVOC conversion and further refinement • Extensive testing on scalability, concurrent use of the Workbench etc. • Create a network of ontology experts • Multilingual subject experts: term editors, ontology editors, validators • Workshops/Trainings • NeOn
The Semantic Web meets WEB 2.0: cataloguing and annotating Web Resources (1/3) • Producing semantically rich, multilingual knowledge repositories • Combine social bookmarking with Ontology based bookmarking • Composing, organizing and offering information from different knowledge repositories (distributed ontologies)
The Semantic Web meets WEB 2.0 (2/3) • Collaborative aspects • Distribution of ontological data • Collaborative semantic annotation • Collaborative editing of ontologies • Data Integration • Ontology Learning • Automatic discovery of ontologies (Ontology Crawling) • Linking/Merging of ontologies • Linguistic Aspects • Identity of concepts: discovering of linguistic phenomena like synonymy/homonymy • Linguistic Enrichment of Ontologies • Multilinguality
The Semantic Web meets WEB 2.0 (3/3) • People/Expert discovery service • Working on the case study: avian flu • Partners • FAO • University of Rome Tor Vergata, Italy • Kasetsart University, Bangkok, Thailand • Yahoo, UK/Ireland • Elsevier, Netherlands • University of Bielefeld, Germany • ....
Semantic Turkey Università Tor Vergata Roma, Italy • Semantic Web technologies are offering standards for knowledge representation and for content exchange • Obtain a clear separation between pure knowledge data (the WHAT) and web links (the WHERE) • Offer innovative navigation of both the acquired information and of the pages where it has been collected • The project has been focused on innovative solutions for browsing the web and for organizing the information which is observed during navigation
OntoLing Università Tor Vergata Roma, Italy • Exploiting Linguistic Resources for building linguistically motivated ontologies • Linguistic Enrichment of Ontologies • Manually • Automatically
Early warning systems (1/3) • Emerging infectious diseases warning system • Use cases: Bird Flu, Rice • Problem: today systems in many domains already exist but there is no efficient access to them. The information is scattered and not organized • Structured information (e.g. the CS) can be used to organize unstructured information (wikis and blogs)
Early warning systems (2/3) • Resources • FAO CS, Metadata and multilingual subject ontologies • AGRIS data • Department of Livestock data • Topics • From unstructured information to structured information • Distributed and collaborative knowledge building • Knowledge mapping • Topic map • Knowledge engineering, knowledge mining, data mining • Intelligent decision support systems (diagnosis analysis)
Early warning systems (3/3) • Partners • Thai Ministry of Science and Technology (NECTEC) • Thai Ministry of Agriculture (Department of Livestock) • Kasetsart University, Bangkok, Thailand • Thai Agris Center, Bangkok, Thailand • HAII, Bangkok, Thailand • FAO • Consortium of Asian countries (APEC) • .....
Grid Technologies in Digital Libraries (1/2) • Combine distributed repositories through ontologies by means of the grid technology • Several institutions • Repositories of scientific data, GIS, statistical data • Domain of fisheries / aquaculture
Grid Technologies in Digital Libraries (2/2) • Partners • FAO • Italian National Research Center (CNR), Italy • Thai National Grid Center, Kasetsart University, Bangkok • Thai National Agris Center, Kasetsart University, Bangkok • NECTEC, Thailand • ....
Thank you! More information: http://www.fao.org/aims/ margherita.sini@fao.org … open discussion …