Raphael Troncy

If this is still needed, DM me as well. I should be able to connect to people with the appropriate skills.

Checkout the Text2Graph product from LettrIA! https://www.lettria.com/fr/features/text-to-graph

Yes, this is what I had in mind

You mean, expressing metadata that will describe the type and volume of data that has been used for training the model, the model architecture, the hyperparameters, etc.? In a structured way? There are numerous initiatives indeed, even ML Commons is a start.

Is https://2024.eswc-conferences.org/wp-content/uploads/2024/04/146640512.pdf what you’re looking for?

Thomas T. SPARQL Transformer DOES expose a RESTful API on top of any RDF graph. Hence, you do not bother anymore with SPARQL. You only do GET / POST API calls in your app. SPARQL Transformer is enabling you to specify the enveloppe response that the developer wishes to have to avoid merging results, deduplicating, etc. It is integrated into grlc so that you can drop SPARQL Transformers template in a git repository and you get immediately for free a RESTful API deployed + Swagger documentation! See http://bit.ly/transformer-iswc19 for all features. More at https://github.com/D2KLab/sparql-transformer + playground at https://d2klab.github.io/sparql-transformer/

Did you check out SPARQL Transformer that does make RDF very developper friendly? There are large RDF-based applications that makes use of it efficiently.

One essential piece is https://github.com/D2KLab/sparql-transformer !