Files in this item



application/vnd.openxmlformats-officedocument.presentationml.presentation2018-10-31--Linh-Hoang--Research-Showcase.pptx (295kB)
(no description provided)Microsoft PowerPoint 2007


Title:Extracting claim sentences from biomedical documents: a pilot study focusing on drug-drug interaction claims
Author(s):Hoang, Linh K.; Boyce, Richard D.; Schneider, Jodi
Subject(s):Text Mining
Claim Mining
Biomedical Informatics
Drug-Drug Interaction
Knowledge Discovery
Abstract:A vast amount of biomedical research is published every year, making it extremely difficult to keep track of new findings. It is possible to represent the main findings of a biomedical paper as “claims” supported by evidence. The long-term goal of this research is “claim mining”, in which we seek to identify and understand what claims are present in a biomedical paper and how the paper’s data and methods support the claims. Towards that objective, we are exploring how claims are represented in biomedical documents and how we can identify them more quickly and efficiently. In this pilot study, we examine claims about drug-drug interactions. Drug-drug interaction claims are a good starting point for research because drug-drug interactions are important for medical treatment, there are over 150,000 PubMed articles indexed with the MeSH term "Drug Interactions", and there exist a variety of supporting resources including drugs databases, drug-drug interactions detection algorithms, and an annotated corpus of drug-drug interactions evidence. Our pilot study seeks to develop a system that, given a biomedical document, automatically identifies sentences that contain drug-drug interaction claims.
Issue Date:2018-10-31
Citation Info:Hoang LK, Boyce, RD, Schneider J. Extracting claim sentences from biomedical documents: a pilot study focusing on drug-drug interaction claims. In The iSchool at Illinois 2018 Research Showcase; Illinois. 2018.
Conference Poster
Date Available in IDEALS:2018-11-25

This item appears in the following Collection(s)

Item Statistics