Use Case AGROVOC Thesaurus
Back to Use Cases & Case Studies page
Food and Agricultural Organization of the United Nations Contact person: Johannes Keizer <johannes.keizer @fao.org>
Background and Current Practice
AGROVOC was developed in the 1980s as a multilingual structured thesaurus for subject fields in agriculture, forestry, fisheries, food and related domains (e.g. environment). Its purpose was to standardize the indexing process for the AGRIS bibliographic database in order to make searching simple, efficient and relevant. With the rise of semantic technologies AGROVOC has been transformed from a simple term based thesaurus into a concept scheme. The use of AGROVOC has also expanded considerably to the point where it is now a model for the organization of explicit knowledge as well as a resource used to add multilingual search functionality. The first version of the AGROVOC concept scheme was developed in OWL and recently converted to a standard SKOS-XL model (Figure 1 and 2). AGROVOC is maintained using a bespoke application known as the Agrovoc Concept Server that allows for distributed maintenance of this global, multilingual resource.
Figure 1. Representation of the data type and object type properties of AGROVOC
agrovoc:c_1344 a skos:Concept; skos:inScheme agrovoc:agrovocScheme; skos:broader agrovoc:c_330829; skosxl:prefLabel agrovoc:xl_en_1278479065260, agrovoc:xl_fr_1278479065347, ... skosxl:altLabel agrovoc:xl_ko_1278479066170, ... agrovoc:xl_en_1278479065260 skosxl:literalForm "cartography"@en; a skosxl:Label; aosc:hasSynonym agrovoc:xl_en_1278479067192, ... agrovoc:xl_fr_1278479065347 skosxl:literalForm "Cartographie"@fr; a skosxl:Label . agrovoc:xl_en_1278479067192 skosxl:literalForm "Map making"@en; a skosxl:Label; aosc:hasSynonym agrovoc:xl_en_1278479065260 ...
Figure 2. A sample output for AGROVOC SKOS-XL(Turtle) .
1. What is achieved in the scenario without reference to linked data?
- Relation of multilingual resources.
- Relation of resources in separate data silos that use AGROVOC for indexing.
- Standardized semantic markup of unstructured resources.
- Aggregation of resources using broader-term relationships.
- Relation of resources using related-term relationships.
- Semantic search
2. How we use linked data technology to achieve this goal?
- We make it easier to express alignments with other agricultural thesauri in an interoperable and reusable way.
- We establish relations between AGROVOC and other agricultural vocabularies in the cloud.
- We encourage information systems that use AGROVOC to get published as Linked Data.
- Information management specialists
- Document repositories
- Library catalogs
- Service providers
Use Case Scenario
Azizallah, an indexer, is creating a bibliographic record for a document on Oreochromis niloticus aquaculture production techniques written by Egypt’s fisheries ministry. He opens his browser and navigates to AGROVOC’s Agrotagger service. He uploads the document and the service returns a list of 7 terms in Arabic. He selects 4 as relevant and rejects the other 3. He notices that the document had referred to the fish as Nilotica and that Agrotagger has not suggested it as a species name. He searches by scientific name Oreochromis niloticus and the system returns it together with a list of synonyms such as mango fish, pomegranate fish and tilapia, all names coming from the FAO Fisheries and Aquaculture Department’s taxonomic database that are associated to AGROVOC. He selects the name. While browsing Tilapia he also notices that cage culture is a related term and he selects that as well. He clicks <output> and the system generates a file that Azizallah pastes into his record.
- Linked data facilitates vocabulary federation, even for vocabularies not yet published as linked data but who map to those who are.
- Linked data using standard markup (e.g. RDF) and standard interfaces (e.g. SPRQL) create services which are easy to exploit thus fostering their use and helping to make linked data vocabularies and thesauri core hubs for connecting other reference vocabularies and data.
- Linked data improves resource discovery by creating explicit relationships within and between concepts and data.
- Linked data facilitates navigation of related concepts and data.
- The AGROVOC Concept Server Work Bench  and AGROVOC WebServices  both contain the AGROVOC Concept Scheme. This scheme in both OWL and SKOS-XL can be used as an indexing mechanism.
- EUROVOC Linked data .
- List (a sample)of the namespaces for AGROVOC Concept Scheme:
Classes / relationships
- OWL 
- SKOS 
- DC and DCTerms 
- ASFA Thesaurus 
- Fisheries Reference Metadata 
- USDA National Agricultural Library 
Problems and Limitations
- Due to legacy existing data there are inconsistencies. In consequence, there is no valid example for definition and image in AGOVOC SKOS-XL yet.
- More semantic enrichment is necessary for KOS integration.
- It may be difficult to federate linked and non-linked thesauri.
- Concept scheme evolution is challenging considering that even deprecated concepts and relationships must maintain an accessible URL.
Library Linked Data Dimensions / Topics
- User needs>Browse/explore/select/retrieve/find
- System> Agris Search Engine> Indexing
- System> Library system>Catalogs
- Context > Communication > Online access and Navigations
- Conceptual Scheme> Knowledge model
- Confirm the use of SKOS for multilingual KOS vocabularies.
- Using SKOS to describe knowledge organization systems (KOS) that have been used by library, archive, and museum communities (LCSH, RAMEAU, AAT, RGN, TGM I and II, DDC, etc.)
Related Use Cases
- This Use Case, Use Case AGRIS, and Use Case FAO Authority Description Concept Scheme are related as follows: