EnrichEuropeana+ will continue the development of the Transcribathon Platform (www.transcribathon.eu), making it available to Cultural Heritage Institutions for engaging public users and professionals in enhancing the semantic and multilingual description of Cultural Heritage objects. The project will develop new services based on innovative AI solutions in order to automate transcription and enrichment activities.
There is a wealth of information about 19th century history laying mostly dormant in archives. Most of the original sources are available only in handwritten form and can not be accessed on a large scale. A new Citizen Science crowdsourcing campaign will be organized to stimulate user engagement for transcribing, translating, adding semantic enrichments and increasing the accessibility of manuscripts related to historical events and societal transformations in Europe within the 19th Century. As a result, these documents will be made available to support
education and research activities.
In order to achieve its goals, the project will aggregate new content to Europeana using the existing national aggregation infrastructures. The crowdsourced enrichments will be submitted back to the original collections and ingested to Europeana increasing the quality of record metadata (Tier level). The already existing Data Exchange Infrastructure and National Aggregator infrastructures will be further developed to create the required functionality to support this operational workflow.
Innovative AI solutions will be employed to automate the transcription and semantic enrichment activities. The Handwritten Text Recognition (HTR) technology developed within the scope of the H2020 Project READ will be integrated to support and reduce the effort of manuscript transcription. Natural language processing and big data analysis technology will be used to analyse transcriptions and their translations, providing support for semantic metadata enrichment, clustering and classification.
The project aims at increasing the scalability and functionality of the Transcribathon tool to transform it into a service used by Cultural Heritage Institutions to crowdsource the enrichment of cultural object descriptions and improve the multilingualism of the metadata. A new crowdsourcing campaign will be organised within the scope of this action focusing on carefully selected datasets which provide context for historical events from the 19th Century and highlight the resulting societal transformations. The results of the campaign will be used to prepare editorial materials in the form of exhibitions and blog posts created in collaboration with Europeana Education and Research Communities.
EnrichEuropeana+ project will be implemented by a consortium of 9 partners. Among them there are 4 institutions that worked on the first edition of the project. Those are Europeana, AIT Austrian Institute of Technology, Facts & Files and PSNC. New partners taking part in the project are READ-COOP, The Provost, Fellows, Foundation Scholars and the other members of Board, of the College of the Holy and Undivided Trinity of Queen Elizabeth near Dublin, Dublin City Council, University of Wrocław and State Archives in Zagreb.