What is linked data in cataloging?

Linked data is a way of publishing structured data that allows metadata from different sources to be interconnected and queried. In the library cataloging context, linked data enables bibliographic records to be linked with authority records and other knowledge organization systems. Some key things to know about linked data in cataloging:

What is linked data?

Linked data refers to a set of best practices for publishing and interlinking structured data on the web. The basic principles of linked data are:

1. Use URIs to identify things
2. Use HTTP URIs so things can be looked up
3. Provide useful information about the thing when its URI is dereferenced
4. Include links to other related URIs

By following these principles, data from different sources can be interconnected through the use of URIs that identify the same real-world object or concept. This enables the web to become a global database, known as the “web of data.”

Why use linked data for library cataloging?

There are several benefits to using linked data for bibliographic records and authority data in library catalogs:

• Connects bibliographic data to other datasets – Linked data allows catalog records to link out to related authoritative datasets like VIAF for authors, LCSH for subjects, etc.
• Improves discovery – By interlinking datasets, relevant information can be pulled together from disparate sources to enhance discovery in the catalog.
• Facilitates data integration – Linked data provides a flexible framework for integrating bibliographic metadata from different libraries and sources.
• Supports semantically richer queries – Linked data enables more intelligent searches across datasets using semantic web technologies.
• Future-proofs catalog data – Linked data positions library metadata to take advantage of new technologies and use cases on the web of data.

What linked data standards are used in libraries?

Several linked data standards have been widely adopted by libraries:

• BIBFRAME – A linked data model for bibliographic description that builds on MARC and integrates library data with the web.
• RDA – The cataloging standard RDA provides guidelines for creating metadata based on linked data principles.
• Schema.org – A common metadata schema that makes library data more readable by web search engines.
• Dublin Core – The Dublin Core metadata standard provides a basic set of interoperable terms for linked bibliographic data.

Other common linked data vocabularies used in libraries include SKOS, FOAF, and OWL.

How are entities identified in linked data?

In linked data, real-world entities like people, places, topics, etc. are identified using HTTP URIs. Some examples:

• Authors – Identified by a URI from VIAF (Virtual International Authority File)
• Subjects – Identified by a URI from LCSH (Library of Congress Subject Headings)
• Works – Identified by a URI from identifiers.org
• Places – Identified by a URI from GeoNames

These URIs act as a unique ID that can be used to declare that two datasets are referring to the same entity. This allows information from different sources to be connected.

How are relationships expressed in linked data?

Linked data uses RDF (Resource Description Framework) to represent relationships between entities. RDF expresses data as “triples” – subject, predicate, object. For example:

By stringing together these triples, detailed relationships can be modeled between resources in linked data.

What are some linked data projects in libraries?

Here are a few notable linked data initiatives from libraries and cultural heritage organizations:

• OCLC’s Linked Data Platform – ExposesLIBRIS, the OCLC Online Union Catalog, Quick Start, and other databases as linked data via a cloud-based platform.
• Europeana – The European digital library provides unified access to millions of books, paintings, films, museum objects, and archival records from across Europe’s cultural heritage institutions.
• British Library – The British Library has published various datasets as linked data, including 13 million bibliographic records from their catalog.
• Swedish National Heritage Board – Transformed their National Register of Cultural Heritage into linked open data with over 1.5 million objects.
• BIBFRAME pilot projects – Early adopters like Princeton University and University of California, Davis are testing the use of BIBFRAME for linked data cataloging.

What tools and systems support linked data in libraries?

Some key tools and systems that enable working with linked data in libraries include:

• Triplestores – Databases for storing and querying RDF triples, like Virtuoso.
• RDF parsers – Software libraries that convert between RDF and other formats like MARC, JSON-LD, etc.
• SPARQL endpoints – Allow users to query linked dataset through the SPARQL query language.
• Linked data editors – Desktop tools like OpenRefine and RDF-Mapper assist catalogers in creating and managing RDF data.
• Identity management – Utilities like OpenRefine Reconciliation manage linking local entities to external linked data URIs.
• Linked data discovery tools – Services like Yewno Discover and Zepheira Library.Link Navigator leverage linked data to enhance search and discovery.

What challenges exist with implementing linked data?

Some key challenges that libraries can face when adopting linked data include:

• Cost – Migrating legacy metadata and systems to linked data standards requires significant investment.
• Expertise – Working with linked data often requires specialized skills like RDF, ontologies, and Semantic Web that may be unfamiliar to catalogers.
• Data quality – Metadata quality issues get amplified with linked data’s emphasis on unique entity identification.
• Sustainability – Ongoing linked data maintenance competes with other priorities and requires coordination between departments.
• Interoperability – Differences in cataloging rules, classifications, and identifiers can hinder linking across datasets.

Overcoming these kinds of challenges is key for libraries to fully realize the benefits of linked data.

Conclusion

Linked data has the potential to significantly improve the user experience of library catalogs by connecting bibliographic records to related authoritative sources on the web. By harnessing linked data standards like BIBFRAME, RDA, and Schema.org, libraries can expose their metadata as part of the growing web of data. While implementing linked data still involves challenges, many libraries are already undertaking initiatives in this space and demonstrating real-world benefits. As linked data adoption accelerates, library users will have access to richer, more interconnected catalogs that enhance discovery and fulfill information needs.