Named Entity Recognition using Neural Networks for Clinical Notes

Abstract : Currently, the best performance for Named Entity Recognition in medical notes is obtained by systems based on neural networks. These supervised systems require precise features in order to learn well fitted models from training data, for the purpose of recognizing medical entities like medication and Adverse Drug Events (ADE). Because it is an important issue before training the neural network, we focus our work on building comprehensive word representations (the input of the neural network), using character-based word representations and word representations. The proposed representation improves the performance of the baseline LSTM. However, it does not reach the performances of the top performing contenders in the challenge for detecting medical entities from clinical notes.
Liste complète des métadonnées

Cited literature [17 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01786995
Contributor : Michel Riveill <>
Submitted on : Monday, May 7, 2018 - 10:38:53 AM
Last modification on : Monday, November 5, 2018 - 3:52:10 PM
Document(s) archivé(s) le : Tuesday, September 25, 2018 - 5:33:59 PM

File

UCA-I3S_2.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01786995, version 1

Citation

Edson Florez, Frédéric Precioso, Romaric Pighetti, Michel Riveill. Named Entity Recognition using Neural Networks for Clinical Notes. NLP Challenges for Detecting Medication and Adverse Drug Events from Electronic Health Records (MADE1.0), University of Massachusetts Lowell, Worcester, Amherst, May 2018, Massachusetts, United States. ⟨hal-01786995⟩

Share

Metrics

Record views

179

Files downloads

346