Digital Cultural Heritage Ontologies

The field of digital cultural heritage ontologies is one of the most challenging ones in information management. The digital cultural heritage field is widely addressed, both to public and researchers, and has a lot of heterogeneity in terms of content and potential uses. It is also a field that -based on the expertise of the library, archives and museums communities- has recorded some of the most important initiatives for information management that all exhibit an insightful and thorough view. This talk will provide an overview of the state of the art developments in the field and will discuss some of the results we see in widespread information systems.

Transcript

1. Digital Cultural Heritage Ontologies Giannis Tsakonas Library & Information Center, University of Patras, Greece Joint SIMDAS & VI-SEEM Workshop “Data management and semantic structures for cross-disciplinary research in the SEEM Region” 25-27 June 2018, Nicosia, Cyprus
2. Cultural heritage • A heterogeneous field. • Immovable objects, e.g. natural spaces, sacred sites, etc. • Archaeological findings, e.g. sculptures, weapons, ceramics, etc. • Ethnological objects, e.g. costumes, popular/daily life objects, etc. • Texts, on various surfaces, in different languages, etc. • By period • Distant (physical forms) • Modern (physical forms) • Contemporary (digital forms, history marks present, but unrecorded, etc.)
3. Cultural heritage information • Cultural Heritage information mostly emphasizes on documentation. • Documentation is based on evidence. • Each object, treated by a different organization, has different levels of exhaustiveness of documentation. • Collection: who, when, how much, etc. • Management: info about conservation, exhibition, transfer, etc. • Research and Documentation: description, information enhancement • Presentation: index/non-index fields, public/private information
4. Digital Cultural Heritage information • Work by: Libraries, Archives and Museums • Libraries: Dublin Core, METS, IFLA LRM • Archives: EAD-Encoded Archival Description • Museums: CIDOC Conceptual Reference Model, LIDO, VRA • Different approaches • Object centered approach • Event centered approach
5. Digital Cultural Heritage information • Mostly interested in exposing relations between parts of the history • Evidence might change; documentation should address changes, which leads to increased care for provenance of information
6. Ontologies in DCH • Documentation • Simple: a thorough recording of evidence • Rich semantics (in comparison to metadata schemas) • Bigger span of coverage (new evidence to be recorded) • Classes of entities than fields/placeholders for information • Integration • Simple: a commonly agreed understanding of evidence • Mapping: align information from various metadata schemas • Mediation: translation from one metadata schema to another
7. Reusing DCH information • An important armory of semantic tools, glued by semantic languages • Allow tailoring data models to specific -operational- demands • “Ontology application profiles” based on formal ontologies • CRMinf Argumentation model • CRMsci Scientific Observation Model • ... • Use to construct metadata schemas • EDM • CARARE
8. CIDOC-CRM • ISO 21127:20/2014. Information and documentation -- A reference ontology for the interchange of cultural heritage information • Factual documentation processes; it does not guide. • Entities (90 classes) linked with Properties (169 relations) Property id Property Name Entity – Domain Entity - Range P1 is identified by (identifies) E1 CRM Entity E41 Appellation • Harmonized with other formal ontologies, see FRBRoo with FRBR
9. Europeana Data Model • Reusing elements from: • OAI Object Reuse & Exchange (ORE) • Simple Knowledge Organization System (SKOS) • Dublin Core • W3C Data Catalog Vocabulary (DCAT) • Creative Commons (CC) • SIOC Services Ontology Module • Compatible with CIDOC • Richer metadata; allow many records for the same object; allow records for compound CH objects
10. CARARE • Reusing elements from: • CIDOC CRM • LIDO • MIDAS Heritage • POLIS • Dublin Core and ESE • Harmonization hub for European immovable cultural objects content • Aggregating resources to feed Europeana • Grouping info: Heritage asset identification, Digital resource, Relations, Activity, Rights and Record information
11. Can DCH ontologies help? • Formal ontologies guide data models and frame their operation. • Define the Cardinality, Mandate and Repetition properties of information. • Define structural properties: in CARARE naming conventions were carried by the Appellation field, similar to CIDOC.
12. Thank you for your attention.