Please use this identifier to cite or link to this item: http://hdl.handle.net/11366/545
Title: Data quality assessment and improvement: a Vrije Universiteit Brussel case study
Authors: Van den Berghe, Steven 
Van Gaeveren, Kyle 
Keywords: data quality assessment
data cleaning
data quality improvement
research data management
Issue Date: 9-Jun-2016
Publisher: euroCRIS
Source: "Communicating and Measuring Research Responsibly: Profiling, Metrics, Impact, Interoperability": Proceedings of the 13th International Conference on Current Research Information Systems (2016) Procedia Computer Science (2016, In Press)
Series/Report no.: CRIS2016: 13th International Conference on Current Research Information Systems (St Andrews, June 9-11, 2016)
Conference: CRIS2016 – St Andrews 
Abstract: As the Vrije Universiteit Brussel switched from an in-house built CRIS to Pure, a large number of data quality issues were discovered. In order to solve these, a large-scale data quality assessment and improvement program was started. The assessment sought to find data quality issues and prioritize cleaning tasks along different dimensions, such as reusability and complexity, while taking into account compliance and stakeholder happiness. Moreover, in doing these assessments, an attempt was made to isolate relatively easy to clean parts of the data in order to make them more feasible for people with less domain-knowledge. Finally, some of these data quality improvement operations turned out to be straightforward enough to fully automate them.
Description: Delivered at the CRIS2016 Conference in St Andrews; published in Procedia Computer Science xx (Jul 2016). Contains conference paper (7 pages) and presentation (pending).
URI: http://hdl.handle.net/11366/545
Appears in Collections:Conference

Files in This Item:
File Description SizeFormat 
Procedia_CRIS2016_VandenBerghe_VanGaeveren.pdf95.21 kBAdobe PDFView/Open
Show full item record

Page view(s)

205
checked on Mar 25, 2017

Download(s)

107
checked on Mar 25, 2017

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.