An exploratory approach to the SympTEMIST challenge

Abstract

This paper describes the approach of the HIBA team in the SympTEMIST shared tasks at the BioCreative VIII challenge and workshop held at the Amia 2023 Symposium. This challenge provides an annotated corpus that enables different systems to evaluate their predictions on the detection of "symptom" mentions in electronic health records and their normalization to SNOMED-CT codes. We submitted our predictions for the entity mention detection using a transformer-based approach, obtaining an F1 measure of 0.71.

This article is part of the Proceedings of the BioCreative VIII Challenge and Workshop: Curation and Evaluation in the era of Generative Models.