Document Type
Poster
Original Publication Date
2017
Journal/Book/Conference Title
AMIA 2017 Annual Symposium
Date of Submission
October 2018
Abstract
The UMLS::Association CUICollector module identifies UMLS Concept Unique Identifier bigrams and their frequencies in a biomedical text corpus. CUICollector was re-implemented in Hadoop MapReduce to improve algorithm speed, flexibility, and scalability. Evaluation of the Hadoop implementation compared to the serial module produced equivalent results and achieved a 28x speedup on a single-node Hadoop system.
Rights
© The Authors
Is Part Of
VCU Computer Science Publications
Included in
Computer Engineering Commons, Databases and Information Systems Commons, Other Computer Sciences Commons
Comments
Poster presented by Amy Olex at AMIA 2017 Annual Symposium, Nov 4-8, 2017.