Controlled vocabularies as Linked Data on the Web

1
Controlled vocabularies as Linked Data on the Web

• Rebecca Guenther
• Network Development MARC Standards Office,
• Library of Congress
• rgue_at_loc.gov

Linked Data program July 13, 2009
2
Outline of presentation

• Types of controlled vocabularies
• Encoding formats for controlled vocabularies
• What is SKOS?
• id.loc.gov vocabulary services
• Example of concept scheme in SKOS ISO 639-2
• Next steps

3
Credits

• Ed Summers, Office of Strategic Initiatives
  leading developer and creator of LCSH in SKOS
• Clay Redding, Network Development MARC
  Standards Office, for developing web service and
  work on other controlled vocabularies

4
Why establish controlled vocabularies?

• Control values that occur in metadata
• Reduce ambiguity
• Control synonyms
• Make documentation available for reuse
• Test and validate terms
• Establish formal relationships among values where
  appropriate

5
Types of Controlled Vocabularies used in metadata
standards

• Lists of enumerated values
• Code lists (e.g. language, country)
• Taxonomies
• Formal Thesauri
• Locally controlled enumerated lists

6
Enumerated lists

• Simple list of terms used in a pull-down menu or
  Web site pick list
• Values enumerated in an XML schema
• Little additional information or structure about
  each value
• Examples
• Enumerated value MD5 for METS CHECKSUMTYPE
• Enumerated value born digital in MODS
  digitalOrigin
• Code and value from a MARC 21 fixed field, e.g.
  code e in Leader/06 is cartographic material

7
Code lists

• Some established as ISO standards and used
  worldwide in many communities for many purposes
• The standard generally standardizes the code, not
  a particular name for it
• Codes are used as identifiers
• Some examples
• ISO 639 family (language codes)
• MARC relator codes
• MARC country codes
• ISO 3166 country codes

8
Thesauri

• A thesaurus is a controlled vocabulary with
  multiple types of relationships
• Example
• Rice
• UF Paddy
• BT Cereals
• BT Plant products
• NT Brown rice
• RT Rice straw

9
Standards maintained at LC contain controlled
vocabularies

• LCSH/NAF
• Thesaurus of Graphic Materials
• ISO 639-2 (language codes)
• MARC (including code lists)
• MODS
• METS
• PREMIS
• MIX (XML schema for NISO Z39.87 Technical
  metadata for digital still images)
• and some others

10
Representing information about controlled
vocabulary values

• Data elements in metadata formats, e.g. MARC
  Authority format
• XML schemas (sometimes as enumeration values)
• RDF/XML and RDFS (Resource Description Framework)
• SKOS
• MADS (Metadata Authority Description Schema)

11
About SKOS

• Simple Knowledge Organization System
• RDF application used to express knowledge
  organization systems such as classifications,
  thesauri, taxonomies, and the concepts within
• Allows distributed, decentralized management of
  KOS through Linked Data-inspired application.
• All concepts and schemes require a URI

12
The SKOS data model (Classes)

• ConceptSchemes (e.g., published vocabularies,
  thesauri, code lists, etc.)
• Concepts (individual entries or terms within the
  larger vocabulary)
• Collections (logical groupings of Concepts)

12
13
SKOS concepts

• Labeling properties prefLabel, altLabel,
  hiddenLabel, notation
• Annotation properties note, historyNote,
  scopeNote, changeNote, editorialNote, example,
  definition
• Associative properties broader, narrower,
  related, broadMatch, narrowMatch, closeMatch,
  exactMatch, minorMatch, majorMatch (match
  properties go to Concepts in external
  ConceptSchemes)

14
Advantages to using SKOS

• SKOS has a defined element set which is
  particularly relevant for controlled vocabularies
• Relationships between entries in a concept scheme
  can be expressed (broader, narrower, etc.)
• Relationships between entries in different
  concept schemes can be expressed (exactMatch,
  related)
• Having a dereferencable URI for concepts and
  their concept schemes enhances the ability to
  provide web services for consumers of these
  standards

15
Reasons for developing a web service for
vocabularies

• Facilitate development and maintenance process
  for vocabularies
• Make controlled lists openly available
• Provide comprehensive information about
  controlled values
• Experiment with semantic web technologies and
  linked data
• Expose vocabularies to wider communities

16
Introducing id.loc.gov

• Library of Congress Authorities Vocabularies
  service http//id.loc.gov
• Allows both human-oriented and programmatic
  access to LC-promulgated authorities and
  vocabularies.
• First offering is Library of Congress Subject
  Headings, but more to come e.g. Thesaurus of
  Graphic Materials, ISO 639-2, MARC code lists,
  etc.

17
Introducing id.loc.gov

• Offers bulk data downloads in several RDF
  serializations (likely more to come)
• Goals
• Convey a clear policy regarding access, usage,
  distribution
• Provide continuous updates to keep the data sets
  fresh
• Only serves data values authority and vocabulary
  data, not bibliographic

18
Some features of id.loc.gov

• Provides resolvability by assigning RESTful URIs.
  Each vocabulary and data value within it
  possesses a resolvable URI
• Known-label searches use when you know the label
  but not the identifier (e.g. LCCN)
  http//id.loc.gov/authorities/label/orchids,
  http//id.loc.gov/authorities/label/orchidaceae
• Visualizations
• Default serialization is RDFa XHTML, which can
  be transformed by RDFa tools
• Influenced by the Linked Data movement
  implements SKOS, REST, and HTTP content
  negotiation

19
Example ISO 639-2 vocabulary

• One in the family of ISO 639 language coding
  standards
• Has a close relationship with other language
  coding standards (ISO 639-1 and -3, MARC)
• LC is maintenance agency
• The standard is the CODE, not the language name
  multiple names are given

20
ISO 639-2 language code in SKOS
ltrdfDescription rdfabout "http//www.loc.gov/st
andards/registry/vocabulary/iso639-2/por"gt
ltrdftype rdfresource"http//www.
w3.org/2008/05/skos Concept"/gt
ltskosprefLabel xmllang"x-notation"gtporlt/skos
prefLabelgt ltskosaltLabel xmllang"en-Latn"gtP
ortugueselt/skosaltLabelgt ltskosaltLabel
xmllang"fr-Latn"gtportugaislt/skosaltLabelgt
ltskosnotation rdfdatatype"xsstring"gtporlt/sk
osnotationgt ltskosdefinition
xmllang"en-Latn"gtThis Concept has not yet been
defined.lt/skosdefinitiongt ltskosinScheme
rdfresource"http//www.loc.gov/standards/regist
ry/vocabulary/iso639-2"/gt ltvsterm_statusgtstable
lt/vsterm_statusgt ltskoshistoryNote
rdfdatatype"xsdateTime"gt2006-07-19T084154.000
- 0500lt/skoshistoryNotegt ltskosexactMatch
rdfresource "http//www.loc.gov/standards/regis
try/vocabulary/iso639-1/pt"/gt ltskosexactMatch
rdfresource "http//www.loc.gov/standards/regis
try/vocabulary/languages/por"/gt
ltskoschangeNote rdfdatatype"xsdateTime"gt2008
-07- 09T134905.321-0400lt/skoschangeNotegt lt/rdf
Descriptiongt
21
(No Transcript)
22
Next steps

• Advocacy, user feedback etc. for LCSH
• Implement update mechanism for processing changes
  processed from LC CDS
• Expand system to allow more vocabularies and
  Linked Data relationships
• Name authorities

23
Next steps

• MADS OWL Schema to enable identification of
  facets within name and subject authorities
  Aeronautics--Soviet UnionHistory
• Develop mechanisms to output all public
  documentation for controlled vocabularies
  already working for ISO 639-5 (master data is
  SKOS)
• http//www.loc.gov/standards/iso639-5/