DE-BIAS - Detecting and cur(at)ing harmful language in cultural heritage collections

DE-BIAS aims to create a tool that automatically detects offensive terms and provides information on their problematic background, using a vocabulary that will pair biased language with contextual information and suggestions for appropriate terms. The project will foster conversations with minoritised communities and involve them in the co-creation of a vocabulary that allows for self-expressive metadata. The tool will be integrated into Europeana website and made available for use by individual cultural heritage institutions to analyse and detect bias in their own datasets. Finally, the project will develop several capacity-building resources, including an online knowledge base and train-the-trainer course, to ensure the long-term impact and adoption of the project’s results by the cultural heritage sector.

The project is funded under Digital Europe Programme (DIGITAL) of the European Union. 

The project will run for 24 months and is led by DFF - Deutsches Filminstitut & Filmmuseum from Germany. The other members of the consortium are:

1. Europeana Foundation (EF) - The Netherlands

2. Datoptron (DATOPTRON) - Greece

3. European Fashion Heritage Association (EFHA) - Italy

4. Thinkcode (THINKCODE) - Cyprus

5. Michael Culture Association (MCA) - Belgium

6. ECCOM Centro Europeo per l’Organizzazione e il Management Culturale (ECCOM) - Italy

7. Katholieke Universiteit Leuven (KU LEUVEN) - Belgium

8. Stichting Archives Portal Europe Foundation (APEF) - The Netherlands

9. Ministère de la Culture et de la Communication (MCC) - France

10. Stichting Nederlands Instituut Voor Beeld En Geluid (NISV) - The Netherlands