When BERT Fails -- The Limits of EHR Classification
Computation and Language
2022-08-25 v1 Machine Learning
Abstract
Transformers are powerful text representation learners, useful for all kinds of clinical decision support tasks. Although they outperform baselines on readmission prediction, they are not infallible. Here, we look into one such failure case, and report patterns that lead to inferior predictive performance.
Cite
@article{arxiv.2208.10245,
title = {When BERT Fails -- The Limits of EHR Classification},
author = {Augusto Garcia-Agundez and Carsten Eickhoff},
journal= {arXiv preprint arXiv:2208.10245},
year = {2022}
}