Copy of ontology and prototype knowledge base of the IKS system

Disclaimer

The base ontology for the knowledge base is CIDOC CRM in the Erlangen CRM implementation, and its FRBRoo extension based on the FRBR (Functional Requirements for Bibliographic Records) proposed by IFLA. The knowledge base contains processed data obtained from the Digital Libraries Federation (whose sources as listed here), the NUKAT union catalogue, and a sample of 15,000 records from the Mona system used by the National Museum of Poland in Warsaw. The FRBRoo implementation also comes from the Erlangen University. However, it is in an early stage of development, so a few inconsistencies have been locally corrected to better match the original FRBRoo specification.

The provided data is a working copy and may contain errors. The quality of knowledge base data is improving with each development process step.

General data model information

The publication in the knowledge base are described by means of four FRBR levels:

  • Work (the ideal work as created by the author) – instances of F14_Individual_Work or F18_Serial_Work).
  • Expression (the intellectual content of a particular edition of a worka) – instances of F24_Publication_Expression
  • Manifestation (set of Items from the same Expression, grouping information common to all items) – instances of klasy F3_Manifestation_Product_Type
  • Item – instances of F5_Item

Museum objects are represented as instances of E22_Man-Made_Object.

Ontologies in the Knowledge Base

Type hierarchies, thesauri, controlled vocabularies

The E55 CIDOC class is an interface allowing for the re-use of existing vocabularies in the ontology. The following classes have been created to represent external hierarchies in the knowledge base:
Klasa E55 jest interfejsem, za pomocą którego można osadzać istniejące tezaurusy i słownictwo kontrolowane. Utworzone przez nas hierarchie to:

  • types based on the classifications in the Mona system used by the National Museum of Poland in Warsaw
  • other known types [CIDOC]
  • The Kaba subjects in the knowledge base are connected with a richer set of relations than most external types:

    • Broader (P127_has_broader_term) and narrower (P127i_has_narrower_term) term relations, created by parsing the subject headings language grammar, in which “Niemcy — 1056-1106 (Henryk IV).” has a broader term “Niemcy.”
    • See also relations. Those were taken from the contents of KABA records. They are less restrictive than the first type of relations and sometimes contain errors, so it is not possible to use them to form a valid hierarchy. Still, the information is valuable. It is represented in the knowledge base by the similarTo property (from open.vocab.org) and its subproperties: P214_see_also_broader_term, P214i_see_also_narrower_term, P213_see_also_earlier_form, P213i_see_also_later_form

    Knowledge base RDF dump

    • Export format: Notation3
    • The knowledge base contains triples generated based on: 100 000 DLF Records, 100 000 NUKAT records, 15 000 National Museum records. Those records were subject to: mapping, enriching and relation detection.
    • It also contains ontology triples.
    • Generation date: 2012-09-03
    • The knowledge base dump does not contain implicit triples (that could be produced by a reasoner)
    • No of triples: 19 149 139
    • Newly created resources are given URI’s in the the http://dl.psnc.pl/kb/ namespace. This is a temporary solution. In Autumn 2012 we plan to release the official version of the knowledge base, together with a SPARQL endpoint. Resources in that knowledge base will have permanent URI’s in another, target namespace.

    2 thoughts on “Copy of ontology and prototype knowledge base of the IKS system

    1. Witam
      Niestety linki nie działają:

      Forbidden

      You don’t have permission to access /sites/synat-protected/ontology/ecrm_current.owl on this server.

    2. W tej chwili wszystkie linki do słownictwa i ontologii powinny działać – zasoby te są już dostępne na zewnątrz.
      Natomiast jeśli chodzi o zrzut bazy wiedzy, to niezbędne jest uzyskanie zgody na dostęp (który już został Ci przyznany).
      Pozdrawiam

    Leave a Reply

    Your email address will not be published. Required fields are marked *

    *