Creating a large database test bed with typographical errors for record linkage evaluation

Evaluation of record linkage algorithms requires a large database test bed that is representative of the real-world data. We created such a large database that reflects the demographic distribution of a typical population and contains typographical errors commonly made during data entry. This databa...

Full description

Saved in:
Bibliographic Details
Published in:AMIA ... Annual Symposium proceedings 2008-11, p.1153-1153
Main Authors: Theera-Ampornpunt, Nawanan, Kijsanayotin, Boonchai, Speedie, Stuart M
Format: Article
Language:English
Subjects:
Databases, Factual
Forms and Records Control
Information Storage and Retrieval - methods
Medical History Taking - methods
Medical Record Linkage
Medical Records Systems, Computerized
Minnesota
Terminology as Topic
Word Processing
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Evaluation of record linkage algorithms requires a large database test bed that is representative of the real-world data. We created such a large database that reflects the demographic distribution of a typical population and contains typographical errors commonly made during data entry. This database can be used with high confidence as a test bed to evaluate various record linkage algorithms.
ISSN:1559-4076