Virtual International Authority File

The VIAF project was initiated in 1998 by the Library of Congress, the Deutsche Nationalbibliothek, and OCLC, with the goal of creating a shared virtual authority file that would reduce duplication and improve international cooperation. The first prototype was demonstrated in 2003, and the service became publicly available in 2006. In 2012, OCLC assumed full operational responsibility, and the database has since grown to include contributions from over 50 national libraries and other institutions, including the Bibliothèque nationale de France, the British Library, and the Vatican Library.

The system uses a clustering algorithm to match authority records from different sources, linking them to a single VIAF cluster. This process is continuously refined to improve accuracy and coverage, and the data are made available as linked open data, supporting the Semantic Web.

VIAF serves as a central hub for authority data, allowing libraries to share and reuse name and subject headings. It is particularly valuable for cataloging, as it enables libraries to adopt established forms of names without re-creating them, and for researchers, who can use VIAF to disambiguate authors and find all works by a given person across different languages and transliterations.

The database is integrated into many library systems and discovery tools, and it is a key component of the International Standard Name Identifier (ISNI) system, which assigns persistent identifiers to public identities. VIAF data are also used in Wikipedia to link biographical articles to authoritative library records, enhancing the reliability of citations.

VIAF is built on a cluster-based model, where each cluster represents a distinct entity (e.g., a person or organization) and aggregates the authority records from contributing libraries. The clustering algorithm uses a combination of exact and fuzzy matching on names, dates, and other attributes, and it is continually updated as new records are added.

The data are stored in a relational database and exposed through a variety of interfaces, including a web search interface, an SRU (Search/Retrieval via URL) API, and a linked data endpoint that supports RDF and SPARQL queries. Each VIAF cluster is assigned a stable URI, such as http://viaf.org/viaf/123456, which can be used as a persistent identifier in other systems.

While VIAF is widely known for its role in library cataloging, it has several lesser-known facets. For instance, VIAF includes authority records for fictional characters and mythological figures, not just real people. It also supports multiple scripts, including Cyrillic, Arabic, and Chinese, and it handles transliteration variants, making it a global resource.

Another notable aspect is VIAF's role in the Person Data project, which aims to create a comprehensive international register of persons in cultural heritage. Additionally, VIAF has been used in digital humanities research to study historical networks and to enrich biographical databases. The project also collaborates with Wikidata, and many VIAF identifiers are cross-linked with Wikidata items, facilitating data interoperability.

Authority file

A collection of standardized headings for names, subjects, and titles used in library catalogs.

Cluster

A group of authority records that refer to the same entity, merged into a single VIAF record.

Linked open data

A method of publishing structured data so that it can be interlinked and become more useful through semantic queries.

SRU

Search/Retrieval via URL, a standard protocol for querying databases.