Deep Learning in Biomedical Text Mining: Contributions and Challenges

Tanvir Alam*, Sebastian Schmeier

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

6 Citations (Scopus)

Abstract

A large number of biomedical texts are published every day in scientific literature. Finding the relevant and useful information from the massive collection of scientific literature is a challenging task that can be compared to finding needles in the haystack. Biomedical text mining is one of the sophisticated methodologies that leverage the extraction of knowledge from existing biomedical texts automatically. Deep learning (DL) based techniques have rejuvenated this field with huge prospects. In this chapter, we highlighted the contribution of DL based techniques in three specific tasks in the field of biomedical text mining: named-entity recognition, relationship extraction, and question answering. We also discussed the DL based models that are proven to be successful in multiple natural language processing tasks and the related challenges we face using such DL based techniques. We believe DL based methods will play a significant role in the coming years for biomedical text mining.

Original languageEnglish
Title of host publicationLecture Notes in Bioengineering
PublisherSpringer Science and Business Media Deutschland GmbH
Pages169-184
Number of pages16
DOIs
Publication statusPublished - 2021

Publication series

NameLecture Notes in Bioengineering
ISSN (Print)2195-271X
ISSN (Electronic)2195-2728

Keywords

  • Deep learning
  • Named-entity recognition
  • Natural Language Processing
  • Question answering
  • Relationship extraction

Fingerprint

Dive into the research topics of 'Deep Learning in Biomedical Text Mining: Contributions and Challenges'. Together they form a unique fingerprint.

Cite this