Identifying Adverse Drug Events Mentions in Tweets Using Attentive, Collocated, and Aggregated Medical Representation

Xinyan Zhao, Deahan Yu, V.G.Vinod Vydiswaran


Abstract
Identifying mentions of medical concepts in social media is challenging because of high variability in free text. In this paper, we propose a novel neural network architecture, the Collocated LSTM with Attentive Pooling and Aggregated representation (CLAPA), that integrates a bidirectional LSTM model with attention and pooling strategy and utilizes the collocation information from training data to improve the representation of medical concepts. The collocation and aggregation layers improve the model performance on the task of identifying mentions of adverse drug events (ADE) in tweets. Using the dataset made available as part of the workshop shared task, we show that careful selection of neighborhood contexts can help uncover useful local information and improve the overall medical concept representation.
Anthology ID:
W19-3209
Volume:
Proceedings of the Fourth Social Media Mining for Health Applications (#SMM4H) Workshop & Shared Task
Month:
August
Year:
2019
Address:
Florence, Italy
Editors:
Davy Weissenbacher, Graciela Gonzalez-Hernandez
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
62–70
Language:
URL:
https://aclanthology.org/W19-3209
DOI:
10.18653/v1/W19-3209
Bibkey:
Cite (ACL):
Xinyan Zhao, Deahan Yu, and V.G.Vinod Vydiswaran. 2019. Identifying Adverse Drug Events Mentions in Tweets Using Attentive, Collocated, and Aggregated Medical Representation. In Proceedings of the Fourth Social Media Mining for Health Applications (#SMM4H) Workshop & Shared Task, pages 62–70, Florence, Italy. Association for Computational Linguistics.
Cite (Informal):
Identifying Adverse Drug Events Mentions in Tweets Using Attentive, Collocated, and Aggregated Medical Representation (Zhao et al., ACL 2019)
Copy Citation:
PDF:
https://aclanthology.org/W19-3209.pdf
Data
SMM4H