[Paper Review] Data Quality Measures and Data Cleansing for Research Information Systems

Otmane Azeroual, Gunter Saake+1 more|arXiv (Cornell University)|Jan 18, 2019

Data Quality and Management11 references58 citations

TL;DR

This paper presents measures and new data cleansing techniques to improve data quality in Research Information Systems (RIS) and supports reliable decision-making.

ABSTRACT

The collection, transfer and integration of research information into different research Information systems can result in different data errors that can have a variety of negative effects on data quality. In order to detect errors at an early stage and treat them efficiently, it is necessary to determine the clean-up measures and the new techniques of data cleansing for quality improvement in research institutions. Thereby an adequate and reliable basis for decision-making using an RIS is provided, and confidence in a given dataset increased. In this paper, possible measures and the new techniques of data cleansing for improving and increasing the data quality in research information systems will be presented and how these are to be applied to the Research information.

Motivation & Objective

• Identify data quality issues arising from collection, transfer, and integration of research information into RIS.
• Propose measures and techniques for data cleansing to improve data quality in research institutions.
• Explain how improved data quality supports reliable decision-making using RIS.

Proposed method

• Review potential data quality issues in RIS due to collection, transfer, and integration.
• Present measures for data cleansing applicable to RIS data.
• Describe new techniques of data cleansing for quality improvement in research information systems.
• Explain application of these measures to the management of research information.

Experimental results

Research questions

• RQ1 What data quality problems arise when feeding data into Research Information Systems?
• RQ2 What measures can be used to detect and clean such data quality issues?
• RQ3 What new data cleansing techniques improve data quality in RIS?
• RQ4 How do these measures support decision-making using RIS?

Key findings

• The paper discusses possible measures to detect and address data errors in RIS data pipelines.
• It outlines new data cleansing techniques aimed at quality improvement for research information systems.
• The proposed measures are intended to provide an adequate and reliable basis for decision-making using RIS and increase dataset confidence.