Skip to main navigation Skip to search Skip to main content

Improving Adverse Drug Event Extraction with SpanBERT on Different Text Typologies

  • Beatrice Portelli (Corresponding Author)
  • , Daniele Passabì
  • , Edoardo Lenzi
  • , Giuseppe Serra
  • , Enrico Santus
  • , Emmanuele Chersoni

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

Abstract

In recent years, Internet users are reporting Adverse Drug Events (ADE) on social media, blogs and health forums. Because of the large volume of reports, pharmacovigilance is seeking to resort to NLP to monitor these outlets. We propose for the first time the use of the SpanBERT architecture for the task of ADE extraction: this new version of the popular BERT transformer showed improved capabilities with multi-token text spans. We validate our hypothesis with experiments on two datasets (SMM4H and CADEC) with different text typologies (tweets and blog posts), finding that SpanBERT combined with a CRF outperforms all the competitors on both of them.

Original languageEnglish
Title of host publicationAI for Disease Surveillance and Pandemic Intelligence: Intelligent Disease Detection in Action
EditorsArash Shaban-Nejad, Martin Michalowski, Simone Bianco
PublisherSpringer Science and Business Media Deutschland GmbH
Pages87-99
Number of pages13
ISBN (Electronic)9783030930806
ISBN (Print)9783030930790
DOIs
Publication statusPublished - 9 Mar 2022
Event5th International Workshop on Health Intelligence, W3PHAI 2021 held in conjection with 35th AAAI Conference on Artificial Intelligence, AAAI 2021 - Virtual, Online
Duration: 8 Feb 2021 → 9 Feb 2021

Publication series

NameStudies in Computational Intelligence
Volume1013
ISSN (Print)1860-949X
ISSN (Electronic)1860-9503

Conference

Conference5th International Workshop on Health Intelligence, W3PHAI 2021 held in conjection with 35th AAAI Conference on Artificial Intelligence, AAAI 2021
CityVirtual, Online
Period8/02/21 → 9/02/21

Keywords

  • Adverse drug events
  • Digital pharmacovigilance
  • Language models
  • Natural language processing
  • Social media texts

ASJC Scopus subject areas

  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Improving Adverse Drug Event Extraction with SpanBERT on Different Text Typologies'. Together they form a unique fingerprint.

Cite this