Enriching Digital Heritage (pre-workshop)

In the startup phase of the project Enriching Digital Heritage with LLMs and LOD, preceding the related Lorentz workshop, we evaluated several locally run large language models (LLMs) on the text processing tasks named entity recognition and named entity linking, and compared their performances with alternative approaches like Spacy and Dandelion. The software and results related to these evaluations can be found in this Github repository:

• https://github.com/enriching-digital-heritage/eval

Additionally we offered users to process their own texts with the LLMs via a web interface. The software running the demo website is available in this Github repository:

• https://github.com/enriching-digital-heritage/demo