Talp-UPC at eHealth-KD challenge 2019: A joint model with contextual embeddings for clinical information extraction
Document typeConference report
Rights accessOpen Access
Most eHealth entity recognition and relation extraction models tackle the identification of entities and relations with independent specialized models. In this article, we show how a single combined model can exploit the correlation between these two tasks to improve the evaluation score of both, while reducing training and execution time. Our model uses both traditional part-of-speech tagging and dependency-parsing of the documents and state-of-the-art pre-trained Contextual Embeddings as input features. Furthermore, Long-Short Term Memory units are used to model close relationships between words while convolution filters are applied for farther dependencies. Our model was able to get the highest score in all three tasks of IberLEF2019’s eHealth-KD competition. This advantage was specially promising in the relation extraction tasks, in which it outperformed the second best model by a margin of 9.3% in F1 Score.
CitationMedina, S.; Turmo, J. Talp-UPC at eHealth-KD challenge 2019: A joint model with contextual embeddings for clinical information extraction. A: Iberian Languages Evaluation Forum. "Proceedings of the Iberian Languages Evaluation Forum (IberLEF 2019): co-located with 35th Conference of the Spanish Society for Natural Language Processing (SEPLN 2019): Bilbao, Spain, September 24th, 2019". CEUR-WS.org, 2019, p. 78-84.