This new resource firstly offers an expanded reading of the explanatory texts, comprising 5,500 entries on individual works in their Spanish version and almost 2,700 in their English version. This expanded reading has been produced using an NLU motor which recognises and clarifies elements in texts. Using this AI it is possible to extract the principal concepts and subjects covered in the entries on each work. The Gnoss-Sherlock motor used for this purpose includes a knowledge base derived from Wikipedia and, in this case, on the information present on the Museum’s own website, which is already structured through its Knowledge Graph and which provides additional material for ensuring accuracy.
Much of the information required for understanding a work of art lies in a description of it that frequently requires an unusual level of knowledge on the part of the non-expert reader today. It offers today’s readers an explanatory context for the texts which describe the work, thus helping to ensure that they are informative and oriented towards learning.
In addition, and further emphasising this project’s educational and explanatory focus, a multi-layered Timeline has been created that contextualises and expands the Museum’s collections. This Timeline is enriched by the Prado’s Knowledge graph and through structured information from other online sources, specifically Wikidata and Wikipedia, with the aim of providing context for the works and artists in the Museum, from the 12th to the 19th centuries. This background includes historical events, monarchs, governors and leading political figures as well as the artistic, scientific and philosophical output of the relevant period. The configuration of the Timeline allows the user to select up to 5 different layers, one of which always will be that of the Prado, and to decide what types of elements are to appear in each of them: works, rulers, military conflicts, writers, architects, philosophers, etc, as well as the order of the layers, with the Museo del Prado’s always functioning as the base and reference layer.
The Timeline is an example of the application of Linked Open Data Web specifications. The concepts used in the Museum’s website, such as artists and their birth and death dates, will encounter their homologous concepts in other information sources, such as Wikipedia and Wikidata, requesting the related data from them for a specific year, for example. This allows for each work and artist in the Museum to be provided with a very specific context while also giving the collection as a whole an extensive informational background that is exceptionally useful for teaching and learning purposes.
The use of AI for the expanded reading of the collections and the creation of the Timeline based on semantic technology, using a range of external data sources such as Wikipedia as a contextual base, is an innovation on a world-wide level in the museum sector. Through the employment of these resources the Museo Nacional del Prado has reinforced its leading position in the application of the most recent and cutting-edge digital technology for the purpose of disseminating its works and artists, consolidating itself as an international reference for museums and cultural institutions in the use of the principles of the semantic web, the interconnected web and most recently, AI.
In the near future, The Prado Museum proposes the construction of an author’s tool for teachers to create lesson plans based on the timeline. The tool will feature several editing and writing utilities that would make it possible to customize and create timelines for a specific educational purpose in order to be used in class or as teaching guides to prepare the visits to the museum.