@inproceedings{aduragba-etal-2020-sentence,
title = "Sentence Contextual Encoder with {BERT} and {B}i{LSTM} for Automatic Classification with Imbalanced Medication Tweets",
author = "Aduragba, Olanrewaju Tahir and
Yu, Jialin and
Senthilnathan, Gautham and
Crsitea, Alexandra",
booktitle = "Proceedings of the Fifth Social Media Mining for Health Applications Workshop {\&} Shared Task",
month = dec,
year = "2020",
address = "Barcelona, Spain (Online)",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.smm4h-1.31",
pages = "165--167",
abstract = "This paper details the system description and approach used by our team for the SMM4H 2020 competition, Task 1. Task 1 targets the automatic classification of tweets that mention medication. We adapted the standard BERT pretrain-then-fine-tune approach to include an intermediate training stage with a biLSTM architecture neural network acting as a further fine-tuning stage. We were inspired by the effectiveness of within-task further pre-training and sentence encoders. We show that this approach works well for a highly imbalanced dataset. In this case, the positive class is only 0.2{\%} of the entire dataset. Our model performed better in both F1 and precision scores compared to the mean score for all participants in the competition and had a competitive recall score.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="aduragba-etal-2020-sentence">
<titleInfo>
<title>Sentence Contextual Encoder with BERT and BiLSTM for Automatic Classification with Imbalanced Medication Tweets</title>
</titleInfo>
<name type="personal">
<namePart type="given">Olanrewaju</namePart>
<namePart type="given">Tahir</namePart>
<namePart type="family">Aduragba</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jialin</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gautham</namePart>
<namePart type="family">Senthilnathan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexandra</namePart>
<namePart type="family">Crsitea</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-dec</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fifth Social Media Mining for Health Applications Workshop & Shared Task</title>
</titleInfo>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Barcelona, Spain (Online)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper details the system description and approach used by our team for the SMM4H 2020 competition, Task 1. Task 1 targets the automatic classification of tweets that mention medication. We adapted the standard BERT pretrain-then-fine-tune approach to include an intermediate training stage with a biLSTM architecture neural network acting as a further fine-tuning stage. We were inspired by the effectiveness of within-task further pre-training and sentence encoders. We show that this approach works well for a highly imbalanced dataset. In this case, the positive class is only 0.2% of the entire dataset. Our model performed better in both F1 and precision scores compared to the mean score for all participants in the competition and had a competitive recall score.</abstract>
<identifier type="citekey">aduragba-etal-2020-sentence</identifier>
<location>
<url>https://aclanthology.org/2020.smm4h-1.31</url>
</location>
<part>
<date>2020-dec</date>
<extent unit="page">
<start>165</start>
<end>167</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Sentence Contextual Encoder with BERT and BiLSTM for Automatic Classification with Imbalanced Medication Tweets
%A Aduragba, Olanrewaju Tahir
%A Yu, Jialin
%A Senthilnathan, Gautham
%A Crsitea, Alexandra
%S Proceedings of the Fifth Social Media Mining for Health Applications Workshop & Shared Task
%D 2020
%8 dec
%I Association for Computational Linguistics
%C Barcelona, Spain (Online)
%F aduragba-etal-2020-sentence
%X This paper details the system description and approach used by our team for the SMM4H 2020 competition, Task 1. Task 1 targets the automatic classification of tweets that mention medication. We adapted the standard BERT pretrain-then-fine-tune approach to include an intermediate training stage with a biLSTM architecture neural network acting as a further fine-tuning stage. We were inspired by the effectiveness of within-task further pre-training and sentence encoders. We show that this approach works well for a highly imbalanced dataset. In this case, the positive class is only 0.2% of the entire dataset. Our model performed better in both F1 and precision scores compared to the mean score for all participants in the competition and had a competitive recall score.
%U https://aclanthology.org/2020.smm4h-1.31
%P 165-167
Markdown (Informal)
[Sentence Contextual Encoder with BERT and BiLSTM for Automatic Classification with Imbalanced Medication Tweets](https://aclanthology.org/2020.smm4h-1.31) (Aduragba et al., SMM4H 2020)
ACL