Just came across this blog post on Towards Data Science by Katie Morris Claveau: it explains how the team built a dataset and developed a model for detecting Covid symptoms in clinical data, using spaCy and Prodigy. The post also includes some integesting insights into the annotation strategy and working with expert annotators, and it covers the strategy and modivation, including selective sampling for annotation, weak supervision and responsible AI fairness techniques.
Related topics
Topic | Replies | Views | Activity | |
---|---|---|---|---|
Med7 — an information extraction model for clinical natural language processing (built with spaCy & Prodigy) | 12 | 4267 | December 11, 2020 | |
Domain-specific NER project | 1 | 1792 | July 8, 2019 | |
Stuck training some NER models (newbie)
|
2 | 1027 | July 16, 2020 | |
Healthcare NER or Text Classificaiton | 2 | 1701 | August 31, 2018 | |
Demographics Entity Extraction from clinical trail eligibility criteria | 2 | 1018 | April 17, 2020 |