Date of Award
9-7-2017
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Information Science
First Advisor
John Talburt
Abstract
Organizations across many industries from banking to medicine depend on master data that describe their customers, products and services. Data integration and unique representations of master data are supported by Entity Resolution processes to link records for the same master entity. Master data is key to the core operations of these organizations. Unfortunately, many of these organizations do not use a systematic method, or in some cases, use no method for measuring and improving the quality of record linkage provided by Entity Resolution. This research presents an implementation and proposed methodology for routinely and systematically measuring the quality of Entity Resolution results. The systematic measurement of quality is augmented with continued improvement through efficient clerical review methods for improving the quality of the Entity representations generated by Entity Resolution.
Recommended Citation
Pullen, Daniel L., "A System for Stratified Sampling of Entity Resolution Results to Assess and Improve Accuracy with Minimal Clerical Review Effort" (2017). Theses and Dissertations. 779.
https://research.ualr.edu/etd/779
