1 / 14

Data Quality Assurance: Ensuring Reliable Information for Users

Explore the importance of quality control in data communication. Learn about trustworthy data sources, quality checks, and user needs. Understand how to assess and present data confidence levels effectively. Discover methods for quantifying data reliability and implementing a robust quality control system.

ielton
Download Presentation

Data Quality Assurance: Ensuring Reliable Information for Users

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. How to communicate quality and confidence of data? Solfrid Agersten / Per-Ove kjensli

  2. Why do we have QC? • By default we are suspicious … • How trustworthy are the data Dependent on: • The instrument • The weather element to observe • The methods • The QC-system • But – 99.x % of data have good quality!

  3. Who are the user? • I believe all data from Meteorological institute are reliable • I do not care about the quality. I need data now! • I want the possibility to select data based on confidence myself – different data to different use

  4. About quality… • Manual corrected/interpolated data: OK or uncertain? • Automatically corrected/interpolated data: OK or uncertain? • Model data: OK, uncertain or bad? • Set a flag for every quality check Too tight checks, or too slack…?

  5. Quality control flag in operation • Restrict the data published to the user • Uncertain … is that good enough for presentation on Internet? • Simplify the flag for the user in Quality Level • OK • Slightly Uncertain • Very Uncertain • Erroneous

  6. How is the quality information used? http:/yr.no: OK and data flagged as Slightly Uncertain - no information about quality!

  7. eKlima.no • Intention to serve most of the users • Self service of data and quality

  8. Useinfo(2) Quality of Original value => Level no

  9. Useinfo(3) Original corrected => Level no

  10. In future: Possible to quantify the probability for the data to be correct?

  11. How to ensure a good quality control? • Data provider must know the connection between the • qualitycontrolcheck • the flag set • the ‘confidence’ of the data. • User doesn’t care about checks and flags, but the confidence • It is not possible to say anything about the confidence without having correct flags…

  12. What is a correct check? • Dependent of datatype • Dependent of criteria – input parameters • Are the checks independent? • Quantify the reliability of the checks to find weights for the checks. • Find total confidence based on the weights of the checks.

More Related