Entitifying Europeana: Building an ecosystem of networked references for Cultural Objects Hugo Manguinhas, Valentine Charles, Antoine Isaac, Tim Hill| Eur uropeana na F Found undat ion
What at is European ana? a? The Platform for E urope’s Digital Cultural Heritage We aggregate metadata: • From all E U countries • ~3,500 galleries, libraries, archives and museums • More than 53M objects • In about 50 languages • Huge amount of references to places, agents, concepts, time Eur uropeana na aggregat ion n inf nfrast ruc uct ur ure Europeana| | CC BY-SA SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
European ana a Linked Dat at a a St rat at egy Our e efforts and l lines of of w wor ork • Europeana Data Model (EDM) offers a base for linking data • We apply automatic enrichment to link source data to reference data • We encourage data providers to contribute their own vocabularies so that we can benefit from data links made at data providers’ level • We encourage alignment activities between domain vocabularies Significant progress have been made, most of it presented in past SWIB! SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
European ana a Linked Dat at a a St rat at egy A strategy f for E Entities As a cornerstone for our strategy we are building an "Entity Collection" • A service that acts as a centralized point of reference and access to data about contextual entities • Caching and curating data from the wider Linked Open Data cloud • A sort of Europeana "knowledge graph" SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
• • • • • • • European ana a Linked Dat at a a St rat at egy Motiv ivatio ion • Improve user experience S upport better ways of searching and navigating through the collections, eliminating ambiguity and clarifying the meaning of descriptions Adapt better to the language of the user by improving the interlinking of data Brings more context to the objects Alleviates polysemy issues E xpands language coverage Contributes to build a web of data ('knowledge graph') that third parties can use to improve their users' experience SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
The E Ent it it y Colle llect io ion Use C Cases Enrichment of Provider’s Data Crowdsourcing ● A controlled vocabulary to help identify ● Objects can be annotated with references to named references to entities entities ● A controlled vocabulary for client applications Entity Collection Europeana Collections Portal Republication for Re-use ● Findability : users can look for entities, not ● Entities can be republished as an open only records (Entity-Based Search) source to the community ● Understandability : Entity Pages group and present all assertions about an entity ● Exploration : Navigation along relationships becomes possible SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
The E Ent it it y Colle llect io ion What c can it e enable? Entity Pages Semantic auto- completion Semantic and Metadata annotations Entity based facets Food & Drink Project Pundit Annotation Client Google Knowledge Card SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
The E Ent it it y Colle llect io ion How d do we choos ose o our t target v vocabularies? As defined in the recent E uropeana Tech Task Force on enrichment and evaluation (presented last year), we consider the following criteria when selecting a vocabulary: • Properly documented and supported by a community • Technically available on the web according to the Linked Data best practices and recipes • Available under an open licence • Multilingual • Abide to a minimal ontological commitment principle • Apply the best practices and standards for the representation, structure and description of vocabularies • Well-connected internally and externally to other vocabularies (preferably spine vocabularies) SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
The E Ent it it y Colle llect io ion Which target vocabularies are w we u using? For historical reasons, the target vocabularies correspond to the ones being used for S emantic E nrichment (as of November 2016): • Places 214,307 a subset of Ge Geonames, corresponding to places which are part of resources E uropean countries and of some specific feature classes. • Agents 165,008 resources a subset of DBp Bpedia corresponding to most of the instances of dbp:Artist with some exceptions, and integrated from 49 DBpedia language editions. 274 • Concepts resources a subset of DBp Bpedia corresponding to a handful of concepts matching the needs from E uropeana Collections. 2,566 • Time S pans resources The chronological periods from Semi miumT mTime me. SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
The E Ent it it y Colle llect io ion Cont ntribut ution t n to mul ultilingua ual c coverage Entities effectively used to enrich Europeana Objects Entities present in the Entity Collection SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
• • • • • • • The E Ent it it y Colle llect io ion Are t these target vocabularies es en enough? • Not enough coreferencing information to other vocabularies particularly to the ones we receive from data providers (e.g. musical instruments, MIMO) Labels and values are not always accurate and normalized need for better reference data (e.g. VIAF) Missing relevant information e.g. roles and professions Need to expand coverage to other types of entities namely Works and E vents SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
The E Ent it it y Colle llect io ion Challenges Investigate and design strategies for: • Integrating new vocabularies that can further improve • entity descriptions and multilingual coverage (e.g. VIAF) • linking between entities (e.g. Wikidata) • Integrating alignments, in particular: • links between local/domain vocabularies to pivot vocabularies • S upporting manual curation of existing and new entities • Keeping up-to-date the information collected from external sources SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
• • • The E Ent it it y Colle llect io ion Our r roadmap f for t the next y years ✔ • Mint E uropeana UR Is for E ntities and update internal references ✔ Make entity services and data available via an API Make use of the API in the Collections Portal Implement support for new vocabularies and entity types SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
The E Ent it it y Colle llect io ion Alpha r relea ease of our n new Entity API More methods will come, for: Creation, Update and Delete; UR I resolution to E uropeana E ntities SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
The E Ent it it y Colle llect io ion DBpedia resource for “ “Mozart” i in o our d data Inter-linking information… still need to switch references to link to Europeana Entities Preferred labels for 48 languages Coreference links to 6 other datasets (e.g. Freebase, Wikidata) SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
The E Ent it it y Colle llect io ion Entity A API - suggest m method /entity/suggest.json?text=neo&lang=en&rows=6 SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
• • Con onclusion on • A S trategy for E ntities is a “must” for E uropeana There is no “one fits all” vocabulary We have a long way to go… ...but we are making progress SWIB16 16 - Ent nt it ifying ng Eur uropeana na: Bui uilding ng an n ecosyst em of ne net worked referenc nces for Cul ult ur ural Object s CC BY-SA
Thank you! hugo.manguinhas@europeana.eu
Recommend
More recommend
