ARM Data Quality Office Historical Data Quality Report (DQR) Review
Poster PDF
Authors
Kenneth Kehoe — University of Oklahoma
Randy A. Peppler — University of Oklahoma
Sean T. Moore — Orbital ATK Inc.
Justin Monroe — University of Oklahoma
Adam Theisen — Argonne National Laboratory
Category
ARM Infrastructure
Description
Data issues discovered and confirmed by ARM as problems are stored in Data Quality Reports (DQRs) and reported to data users. Tools and processes have been developed in recent years to automatically filter data based on information contained in these DQRs. So, it is now, more important than ever, to ensure that the information in these DQRs is correct and up-to-date. To assist in this matter, the Data Quality (DQ) Office is currently reviewing all historical DQRs and compiling a list of recommended corrections. There are over 7,000 DQRs to review. Of these, an initial 10% have been reviewed by the DQ Office to determine the range of problems that will need to be addressed. Of the 10% reviewed, approximately 65% require revisions or additional review by subject matter experts in order to determine if they are correct. The second phase of this project will be to apply these revisions to the DQR database using information and recommendations from the first phase reviews. We present a plan for completing this review and the application of needed revisions, and propose some tools and methods to help in this effort.Lead PI
Kenneth Kehoe — University of Oklahoma