Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Multilingual challenges for accessing digitized culture online - Riga Summit 15


Published on

"Multilingual challenges for accessing digitized culture online". Presentation at the Riga Summit on the Multilingual Digital Single Market, April 27-29 2015.

Published in: Technology
  • Be the first to comment

  • Be the first to like this

Multilingual challenges for accessing digitized culture online - Riga Summit 15

  1. 1. Multilingual challenges for accessing digitized culture online Antoine Isaac Multilingual Digital Single Market Riga Summit, 28 April 2015
  2. 2. Europe’s platform to access cultural heritage 42M objects and counting
  3. 3. Built on metadata from a broad network Museums National Aggregators Regional AggregatorsArchives Thematic projects Libraries Musées Lausannois The European Library APEX LinkedHeritage Europeana Fashion 2,300 museums, archives and libraries
  4. 4. A platform for access and re-use
  5. 5. Data from all over EU top 16 Items from 36 countries Metadata in 33 languages
  6. 6. Data from all over EU
  7. 7. (Re-)users from all over EU Portal in 31 languages Data access from anywhere
  8. 8. Diversity of domains makes it hard  Data is heterogeneous  Makes it harder to train tools
  9. 9. Diversity of applications makes it harder  Display  Search  Data services ….
  10. 10. Some issues for applying language technology for online culture  Typical queries are very short  Identification of query language is not easy, even manually  Plenty of named entities (persons, places…) Studies by Humboldt University on Europeana and The European Library
  11. 11. Progress and issues – data enrichment
  12. 12. Problems of multilingual enrichment
  13. 13. Problems of multilingual enrichment Poisonous India or the Importance of a Semantic and Multilingual Enrichment Strategy Marlies Olensky, Juliane Stiller, Evelyn Dröge, MTSR 2012 35233-1_25
  14. 14. Exploiting multilingual data from partners
  15. 15. Problems need pooling resources  From cultural heritage sector  From language technology industry  EU level: Horizon2020 projects, DSIs • MT@EU  National level • Cooperation Tilde & Latvian Ministry of Culture
  16. 16. Being open  Open source and open data, as much as possible!  Sharing the word about problems and solutions EuropeanaTech Insight first issue on multilingualism
  17. 17. Thank you! Antoine Isaac @EuropeanaEU