Contextualized Graph Embeddings for Adverse Drug Event Detection

Ya Gao; Shaoxiong Ji; Tongxuan Zhang; Prayag Tiwari; Pekka Marttinen

doi:10.1007/978-3-031-26390-3_35

Contextualized Graph Embeddings for Adverse Drug Event Detection

Ya Gao, Shaoxiong Ji, Tongxuan Zhang, Prayag Tiwari, Pekka Marttinen

Tianjin Normal University

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

2 Citations (Scopus)

142 Downloads (Pure)

Abstract

An adverse drug event (ADE) is defined as an adverse reaction resulting from improper drug use, reported in various documents such as biomedical literature, drug reviews, and user posts on social media. The recent advances in natural language processing techniques have facilitated automated ADE detection from documents. However, the contextualized information and relations among text pieces are less explored. This paper investigates contextualized language models and heterogeneous graph representations. It builds a contextualized graph embedding model for adverse drug event detection. We employ different convolutional graph neural networks and pre-trained contextualized embeddings as the building blocks. Experimental results show that our methods can improve the performance by comparing recent ADE detection models, suggesting that a text graph can capture causal relationships and dependency between different entities in a document.

Original language	English
Title of host publication	Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2022, Proceedings
Subtitle of host publication	European Conference, ECML PKDD 2022, Grenoble, France, September 19–23, 2022, Proceedings, Part II
Editors	Massih-Reza Amini, Stéphane Canu, Asja Fischer, Tias Guns, Petra Kralj Novak, Grigorios Tsoumakas
Publisher	Springer
Pages	605–620
Number of pages	16
ISBN (Electronic)	978-3-031-26390-3
ISBN (Print)	978-3-031-26389-7
DOIs	https://doi.org/10.1007/978-3-031-26390-3_35
Publication status	Published - 17 Mar 2023
MoE publication type	A4 Conference publication
Event	European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases - Grenoble, France Duration: 19 Sept 2022 → 23 Sept 2022 https://2022.ecmlpkdd.org/

Publication series

Name	Lecture notes in computer science
Volume	13714
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases
Abbreviated title	ECML-PKDD
Country/Territory	France
City	Grenoble
Period	19/09/2022 → 23/09/2022
Internet address	https://2022.ecmlpkdd.org/

Access to Document

10.1007/978-3-031-26390-3_35

Contextualized Graph Embeddings for Adverse Drug Event DetectionSubmitted manuscript, 682 KB
SCI_Gao_etal_ECML PKDD_2022Final published version, 483 KB

Cite this

Gao, Y., Ji, S., Zhang, T., Tiwari, P., & Marttinen, P. (2023). Contextualized Graph Embeddings for Adverse Drug Event Detection. In M.-R. Amini, S. Canu, A. Fischer, T. Guns, P. Kralj Novak, & G. Tsoumakas (Eds.), Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2022, Proceedings: European Conference, ECML PKDD 2022, Grenoble, France, September 19–23, 2022, Proceedings, Part II (pp. 605–620). (Lecture notes in computer science; Vol. 13714). Springer. https://doi.org/10.1007/978-3-031-26390-3_35

Gao, Ya ; Ji, Shaoxiong ; Zhang, Tongxuan et al. / Contextualized Graph Embeddings for Adverse Drug Event Detection. Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2022, Proceedings: European Conference, ECML PKDD 2022, Grenoble, France, September 19–23, 2022, Proceedings, Part II. editor / Massih-Reza Amini ; Stéphane Canu ; Asja Fischer ; Tias Guns ; Petra Kralj Novak ; Grigorios Tsoumakas. Springer, 2023. pp. 605–620 (Lecture notes in computer science).

@inproceedings{9f73c6fe770f4acb85cb757345fef310,

title = "Contextualized Graph Embeddings for Adverse Drug Event Detection",

abstract = "An adverse drug event (ADE) is defined as an adverse reaction resulting from improper drug use, reported in various documents such as biomedical literature, drug reviews, and user posts on social media. The recent advances in natural language processing techniques have facilitated automated ADE detection from documents. However, the contextualized information and relations among text pieces are less explored. This paper investigates contextualized language models and heterogeneous graph representations. It builds a contextualized graph embedding model for adverse drug event detection. We employ different convolutional graph neural networks and pre-trained contextualized embeddings as the building blocks. Experimental results show that our methods can improve the performance by comparing recent ADE detection models, suggesting that a text graph can capture causal relationships and dependency between different entities in a document.",

author = "Ya Gao and Shaoxiong Ji and Tongxuan Zhang and Prayag Tiwari and Pekka Marttinen",

note = "| openaire: EC/H2020/101016775/EU//INTERVENE ; European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML-PKDD ; Conference date: 19-09-2022 Through 23-09-2022",

year = "2023",

month = mar,

day = "17",

doi = "10.1007/978-3-031-26390-3_35",

language = "English",

isbn = "978-3-031-26389-7",

series = "Lecture notes in computer science",

publisher = "Springer",

pages = "605–620",

editor = "Massih-Reza Amini and St{\'e}phane Canu and Asja Fischer and Tias Guns and {Kralj Novak}, Petra and Grigorios Tsoumakas",

booktitle = "Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2022, Proceedings",

address = "Germany",

url = "https://2022.ecmlpkdd.org/",

}

Gao, Y, Ji, S, Zhang, T, Tiwari, P & Marttinen, P 2023, Contextualized Graph Embeddings for Adverse Drug Event Detection. in M-R Amini, S Canu, A Fischer, T Guns, P Kralj Novak & G Tsoumakas (eds), Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2022, Proceedings: European Conference, ECML PKDD 2022, Grenoble, France, September 19–23, 2022, Proceedings, Part II. Lecture notes in computer science, vol. 13714, Springer, pp. 605–620, European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Grenoble, France, 19/09/2022. https://doi.org/10.1007/978-3-031-26390-3_35

Contextualized Graph Embeddings for Adverse Drug Event Detection. / Gao, Ya; Ji, Shaoxiong; Zhang, Tongxuan et al.
Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2022, Proceedings: European Conference, ECML PKDD 2022, Grenoble, France, September 19–23, 2022, Proceedings, Part II. ed. / Massih-Reza Amini; Stéphane Canu; Asja Fischer; Tias Guns; Petra Kralj Novak; Grigorios Tsoumakas. Springer, 2023. p. 605–620 (Lecture notes in computer science; Vol. 13714).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Contextualized Graph Embeddings for Adverse Drug Event Detection

AU - Gao, Ya

AU - Ji, Shaoxiong

AU - Zhang, Tongxuan

AU - Tiwari, Prayag

AU - Marttinen, Pekka

N1 - | openaire: EC/H2020/101016775/EU//INTERVENE

PY - 2023/3/17

Y1 - 2023/3/17

N2 - An adverse drug event (ADE) is defined as an adverse reaction resulting from improper drug use, reported in various documents such as biomedical literature, drug reviews, and user posts on social media. The recent advances in natural language processing techniques have facilitated automated ADE detection from documents. However, the contextualized information and relations among text pieces are less explored. This paper investigates contextualized language models and heterogeneous graph representations. It builds a contextualized graph embedding model for adverse drug event detection. We employ different convolutional graph neural networks and pre-trained contextualized embeddings as the building blocks. Experimental results show that our methods can improve the performance by comparing recent ADE detection models, suggesting that a text graph can capture causal relationships and dependency between different entities in a document.

AB - An adverse drug event (ADE) is defined as an adverse reaction resulting from improper drug use, reported in various documents such as biomedical literature, drug reviews, and user posts on social media. The recent advances in natural language processing techniques have facilitated automated ADE detection from documents. However, the contextualized information and relations among text pieces are less explored. This paper investigates contextualized language models and heterogeneous graph representations. It builds a contextualized graph embedding model for adverse drug event detection. We employ different convolutional graph neural networks and pre-trained contextualized embeddings as the building blocks. Experimental results show that our methods can improve the performance by comparing recent ADE detection models, suggesting that a text graph can capture causal relationships and dependency between different entities in a document.

UR - http://www.scopus.com/inward/record.url?scp=85151055046&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-26390-3_35

DO - 10.1007/978-3-031-26390-3_35

M3 - Conference article in proceedings

SN - 978-3-031-26389-7

T3 - Lecture notes in computer science

SP - 605

EP - 620

BT - Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2022, Proceedings

A2 - Amini, Massih-Reza

A2 - Canu, Stéphane

A2 - Fischer, Asja

A2 - Guns, Tias

A2 - Kralj Novak, Petra

A2 - Tsoumakas, Grigorios

PB - Springer

T2 - European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases

Y2 - 19 September 2022 through 23 September 2022

ER -

Gao Y, Ji S, Zhang T, Tiwari P, Marttinen P. Contextualized Graph Embeddings for Adverse Drug Event Detection. In Amini MR, Canu S, Fischer A, Guns T, Kralj Novak P, Tsoumakas G, editors, Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2022, Proceedings: European Conference, ECML PKDD 2022, Grenoble, France, September 19–23, 2022, Proceedings, Part II. Springer. 2023. p. 605–620. (Lecture notes in computer science). doi: 10.1007/978-3-031-26390-3_35

Contextualized Graph Embeddings for Adverse Drug Event Detection

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

INTERVENE: International consortium for integrative genomics prediction

-: Intelligent Crop Production: Data-integrative, Multi-task Learning Meets Crop Simulator

Science-IT

Cite this

Contextualized Graph Embeddings for Adverse Drug Event Detection

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Projects

INTERVENE: International consortium for integrative genomics prediction

-: Intelligent Crop Production: Data-integrative, Multi-task Learning Meets Crop Simulator

Equipment

Science-IT

Cite this