Dependency and AMR embeddings for drug-drug interaction extraction from biomedical literature

Yanshan Wang, Liwei Wang, Sijia Liu, Feichen Shen, Hongfang Liu, Majid Rastegar-Mojarad, Fei Liu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

10 Scopus citations

Abstract

Drug-drug interaction (DDI) is an unexpected change in a drug's effect on the human body when the drug and a second drug are co-prescribed and taken together. As many DDIs are frequently reported in biomedical literature, it is important to mine DDI information from literature to keep DDI knowledge up to date. One of the SemEval challenges in the year 2011 and 2013 was designed to tackle the task where the best system achieved an F1 score of 0.80. In this paper, we propose to utilize dependency embeddings and Abstract Meaning Representation (AMR) embeddings as features for extracting DDIs. Our contribution is two-fold. First, we employed dependency embeddings, previously shown effective for sentence classification, for DDI extraction. The dependency embeddings incorporated structural syntactic contexts into the embeddings, which were not present in the conventional word embeddings. Second, we proposed a novel syntactic embedding approach using AMR. AMR aims to abstract away from syntactic idiosyncrasies and attempts to capture only the core meaning of a sentence, which could potentially improve DDI extraction from sentences. Two classifiers (Support Vector Machine and Random Forest) taking these embedding features as input were evaluated on the DDIExtraction 2013 challenge corpus. The experimental results show the effectiveness of dependency and AMR embeddings in the DDI extraction task. The best performance was obtained by combining word, dependency and AMR embeddings (F1 score=0.84).

Original languageEnglish (US)
Title of host publicationACM-BCB 2017 - Proceedings of the 8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics
PublisherAssociation for Computing Machinery, Inc
Pages36-43
Number of pages8
ISBN (Electronic)9781450347228
DOIs
StatePublished - Aug 20 2017
Event8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics, ACM-BCB 2017 - Boston, United States
Duration: Aug 20 2017Aug 23 2017

Publication series

NameACM-BCB 2017 - Proceedings of the 8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics

Other

Other8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics, ACM-BCB 2017
CountryUnited States
CityBoston
Period8/20/178/23/17

Keywords

  • Abstract Meaning Representation
  • Deep learning
  • Dependency
  • Drug-drug interaction
  • Embeddings

ASJC Scopus subject areas

  • Software
  • Biomedical Engineering
  • Health Informatics
  • Computer Science Applications

Fingerprint Dive into the research topics of 'Dependency and AMR embeddings for drug-drug interaction extraction from biomedical literature'. Together they form a unique fingerprint.

  • Cite this

    Wang, Y., Wang, L., Liu, S., Shen, F., Liu, H., Rastegar-Mojarad, M., & Liu, F. (2017). Dependency and AMR embeddings for drug-drug interaction extraction from biomedical literature. In ACM-BCB 2017 - Proceedings of the 8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics (pp. 36-43). (ACM-BCB 2017 - Proceedings of the 8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics). Association for Computing Machinery, Inc. https://doi.org/10.1145/3107411.3107426