@inproceedings{miftahutdinov-etal-2019-kfu,
title = "{KFU} {NLP} Team at {SMM}4{H} 2019 Tasks: Want to Extract Adverse Drugs Reactions from Tweets? {BERT} to The Rescue",
author = "Miftahutdinov, Zulfat and
Alimova, Ilseyar and
Tutubalina, Elena",
editor = "Weissenbacher, Davy and
Gonzalez-Hernandez, Graciela",
booktitle = "Proceedings of the Fourth Social Media Mining for Health Applications ({\#}SMM4H) Workshop {\&} Shared Task",
month = aug,
year = "2019",
address = "Florence, Italy",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/W19-3207",
doi = "10.18653/v1/W19-3207",
pages = "52--57",
abstract = "This paper describes a system developed for the Social Media Mining for Health (SMM4H) 2019 shared tasks. Specifically, we participated in three tasks. The goals of the first two tasks are to classify whether a tweet contains mentions of adverse drug reactions (ADR) and extract these mentions, respectively. The objective of the third task is to build an end-to-end solution: first, detect ADR mentions and then map these entities to concepts in a controlled vocabulary. We investigate the use of a language representation model BERT trained to obtain semantic representations of social media texts. Our experiments on a dataset of user reviews showed that BERT is superior to state-of-the-art models based on recurrent neural networks. The BERT-based system for Task 1 obtained an F1 of 57.38{\%}, with improvements up to +7.19{\%} F1 over a score averaged across all 43 submissions. The ensemble of neural networks with a voting scheme for named entity recognition ranked first among 9 teams at the SMM4H 2019 Task 2 and obtained a relaxed F1 of 65.8{\%}. The end-to-end model based on BERT for ADR normalization ranked first at the SMM4H 2019 Task 3 and obtained a relaxed F1 of 43.2{\%}.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="miftahutdinov-etal-2019-kfu">
<titleInfo>
<title>KFU NLP Team at SMM4H 2019 Tasks: Want to Extract Adverse Drugs Reactions from Tweets? BERT to The Rescue</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zulfat</namePart>
<namePart type="family">Miftahutdinov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ilseyar</namePart>
<namePart type="family">Alimova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Elena</namePart>
<namePart type="family">Tutubalina</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2019-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fourth Social Media Mining for Health Applications (#SMM4H) Workshop & Shared Task</title>
</titleInfo>
<name type="personal">
<namePart type="given">Davy</namePart>
<namePart type="family">Weissenbacher</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Graciela</namePart>
<namePart type="family">Gonzalez-Hernandez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Florence, Italy</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper describes a system developed for the Social Media Mining for Health (SMM4H) 2019 shared tasks. Specifically, we participated in three tasks. The goals of the first two tasks are to classify whether a tweet contains mentions of adverse drug reactions (ADR) and extract these mentions, respectively. The objective of the third task is to build an end-to-end solution: first, detect ADR mentions and then map these entities to concepts in a controlled vocabulary. We investigate the use of a language representation model BERT trained to obtain semantic representations of social media texts. Our experiments on a dataset of user reviews showed that BERT is superior to state-of-the-art models based on recurrent neural networks. The BERT-based system for Task 1 obtained an F1 of 57.38%, with improvements up to +7.19% F1 over a score averaged across all 43 submissions. The ensemble of neural networks with a voting scheme for named entity recognition ranked first among 9 teams at the SMM4H 2019 Task 2 and obtained a relaxed F1 of 65.8%. The end-to-end model based on BERT for ADR normalization ranked first at the SMM4H 2019 Task 3 and obtained a relaxed F1 of 43.2%.</abstract>
<identifier type="citekey">miftahutdinov-etal-2019-kfu</identifier>
<identifier type="doi">10.18653/v1/W19-3207</identifier>
<location>
<url>https://aclanthology.org/W19-3207</url>
</location>
<part>
<date>2019-08</date>
<extent unit="page">
<start>52</start>
<end>57</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T KFU NLP Team at SMM4H 2019 Tasks: Want to Extract Adverse Drugs Reactions from Tweets? BERT to The Rescue
%A Miftahutdinov, Zulfat
%A Alimova, Ilseyar
%A Tutubalina, Elena
%Y Weissenbacher, Davy
%Y Gonzalez-Hernandez, Graciela
%S Proceedings of the Fourth Social Media Mining for Health Applications (#SMM4H) Workshop & Shared Task
%D 2019
%8 August
%I Association for Computational Linguistics
%C Florence, Italy
%F miftahutdinov-etal-2019-kfu
%X This paper describes a system developed for the Social Media Mining for Health (SMM4H) 2019 shared tasks. Specifically, we participated in three tasks. The goals of the first two tasks are to classify whether a tweet contains mentions of adverse drug reactions (ADR) and extract these mentions, respectively. The objective of the third task is to build an end-to-end solution: first, detect ADR mentions and then map these entities to concepts in a controlled vocabulary. We investigate the use of a language representation model BERT trained to obtain semantic representations of social media texts. Our experiments on a dataset of user reviews showed that BERT is superior to state-of-the-art models based on recurrent neural networks. The BERT-based system for Task 1 obtained an F1 of 57.38%, with improvements up to +7.19% F1 over a score averaged across all 43 submissions. The ensemble of neural networks with a voting scheme for named entity recognition ranked first among 9 teams at the SMM4H 2019 Task 2 and obtained a relaxed F1 of 65.8%. The end-to-end model based on BERT for ADR normalization ranked first at the SMM4H 2019 Task 3 and obtained a relaxed F1 of 43.2%.
%R 10.18653/v1/W19-3207
%U https://aclanthology.org/W19-3207
%U https://doi.org/10.18653/v1/W19-3207
%P 52-57
Markdown (Informal)
[KFU NLP Team at SMM4H 2019 Tasks: Want to Extract Adverse Drugs Reactions from Tweets? BERT to The Rescue](https://aclanthology.org/W19-3207) (Miftahutdinov et al., ACL 2019)
ACL