Data & crawling · Built for a client · 2021–2022

Supplement ingredient dataset

A knowledge base of supplement ingredients and synonyms plus a recogniser that finds them in free-text product listings, for an outside client.

What I did

Replaced the training portal with an API server so training and testing run programmatically.
Chose a Wikibase ontology editor alongside Protege and built the bulk-import deployment.
Specified automated ontology retrieval and a loop feeding trainer classifications back into training.

Write a line; I’ll find it in my inbox.