Date of Award

7-27-2022

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Information Science

First Advisor

John Talburt

Abstract

In an era, where data is being used by organizations in achieving their business goals and driving their business decisions, it is key to ensure the quality of data. For an organization, the most important data assets include master data assets such as product master data and customer master data, supplier master data, employee master data which can be generalized as party master data. There has been significant growth and variety in data in recent years because of which the traditional rules-based approach and dependency on data experts for assessing data quality is no longer working. This dissertation evaluates and documents data quality assessment automation techniques that were available and relevant to the data elements of party and product master data. It also covers which techniques work best in the context of party and product master data and the limitations/dependencies of the techniques. Overall, the evaluation concludes that functional dependencies, machine learning, and fuzzy string match work best for party and product master data quality assessment automation.

Download

Share

COinS