A new training prototype focusing on “AI for Religious Studies - Automatic Keyword Tagging of Multimedia Data” has been delivered by the consortium partner CINECA. Like previous prototypes developed by RESILIENCE, this training serves as a foundation for creating models for future courses and provides valuable insights for enhancing and refining the framework for RESILIENCE training services.
The course was conducted remotely in May 2025 by eight experts from CINECA, a center of excellence in the Italian and European ecosystem of supercomputing technologies for the support and development of pioneering applications. The eight-hour course aimed to give librarians, archivists, collection managers and researchers an understanding of various AI-driven processes, including image enhancement, object recognition and text recognition, to support the valorisation, accessibility and exploration of various types of sources for the study of religion.
The first part of the course focused on AI image enhancement and object detection. Participants learned how to improve the resolution of digital images and detect specific elements within visual digital assets. This approach is used for the handling of resources in the field of research on religion, such as paintings, photographs, manuscripts, as it ensures that valuable artefacts and documents are digitally preserved in high quality and can be accurately identified and categorised.
The second part of the training delved into text recognition technologies, such as Named Entity Recognition (NER), Optical Character Recognition (OCR), and Handwritten Character Recognition (HCR). These tools enable the automatic classification of text content, recognition of printed characters, and transcription of handwritten and typewritten texts. Participants were introduced to open science tools from Hugging Face, which support these procedures.
The training concluded with discussions on the application of the tools and techniques discussed. Examples provided by the participants were used to demonstrate practical applications relevant to their work. By the end of the course, participants had gained valuable insights into using AI to enhance digital images, tag multimedia data, and manage archiving procedures.
The techniques introduced in this training signify a notable advancement in the integration of AI, enhancing both the accessibility and quality of data for the study of religion. As one participant noted in the event evaluation:
“ …at the end of the day I felt completely up to date on the application of AI in Religious Studies.”
See also Prototyping RESILIENCE’s Training Programme – RESILIENCE