|
|
| (19 intermediate revisions by 3 users not shown) |
| Line 6: |
Line 6: |
| | |Sources=https://en.wikipedia.org/wiki/Data_quality; http://iaidq.org/main/glossary.shtml#I | | |Sources=https://en.wikipedia.org/wiki/Data_quality; http://iaidq.org/main/glossary.shtml#I |
| | }} | | }} |
| − | = Data Quality in the oedb =
| |
| | | | |
| − | The quality of all datasets in the '''oedb''' is analysed and rated by defined criteria:<br/>
| |
| | | | |
| − | The first part of the quality criteria is connected to the '''database rules''',<br/>the second part is based on '''geometry correctness''' and<br/>a third part is based on '''user evaluation'''.
| + | |
| | + | = Data Quality and Quality Criteria<br/> = |
| | + | |
| | + | Data, datasets, descriptive data and metadata have to meet "quality criteria, such as accuracy, completeness, consistency, and actuality. [...] Additionally, Open Data credibility and transparency are crucial in order for Open Data to be trustworthy. Origin, originality, and data changes must be traceable and quality features testable." ([http://open-data.fokus.fraunhofer.de/en/projects/open-data-quality/ Fraunhofer FOKUS], see also [https://joinup.ec.europa.eu/sites/default/files/d2.1.2_training_module_2.2_open_data_quality_v1.00_en.pdf here]) |
| | + | |
| | + | |
| | + | |
| | + | |
| | + | = Data Quality in the open energy database (oedb)<br/> = |
| | + | |
| | + | The database set-up of the [[Database|oedb]] is designed to support users in achieving good data quality:<br/> |
| | + | |
| | + | #The interfaces (API and web interface) of the database structure the data and their metadata. In this way the database tries to ensure for instance the submission of all relevant metadata belonging to a dataset. Another example is the automatically generated primary key, which ensures that every row in a table has got an unique identifier. For further information see also [[DatabaseRules|DatabaseRules]]. |
| | + | #Plausibility and integration tests are applied to identify mistakes in the data, e.g. the geometrical correctness test which makes the manual testing with SQL-functions obsolete. |
| | + | #When the number of users becomes large enough, a third level of data testing and rating based on user evaluations might be implemented. |
| | + | |
| | + | Users can contribute to data quality for instance by following the [[DatabaseRules|DatabaseRules]]. For further reading and guidelines with regard to data management and publication see for instance: [https://okfn.org/opendata/how-to-open-data Open Knowledge Foundation], [http://www.odaf.org/?lvl1=resources&lvl2=papers Open Data Foundation] and [http://software-carpentry.org/ Software Carpentry] (e.g. [https://github.com/swcarpentry/good-enough-practices-in-scientific-computing/blob/gh-pages/good-enough-practices-for-scientific-computing.pdf here]). |
| | + | |
| | + | <br/> |
| | + | |
| | + | <br/> |
| | + | |
| | + | <br/> |
Latest revision as of 19:17, 26 July 2016
[edit] Definition
1. Data quality refers to the level of quality of data.
2. The fitness for use of information; information that meets the requirements of its authors, users, and administrators. (Martin Eppler)
[edit] Abbreviation
[edit] Synonyms
Information Quality
[edit] Superterms
Data
[edit] Subterms
[edit] Sources
https://en.wikipedia.org/wiki/Data_quality; http://iaidq.org/main/glossary.shtml#I
[edit] Data Quality and Quality Criteria
Data, datasets, descriptive data and metadata have to meet "quality criteria, such as accuracy, completeness, consistency, and actuality. [...] Additionally, Open Data credibility and transparency are crucial in order for Open Data to be trustworthy. Origin, originality, and data changes must be traceable and quality features testable." (Fraunhofer FOKUS, see also here)
[edit] Data Quality in the open energy database (oedb)
The database set-up of the oedb is designed to support users in achieving good data quality:
- The interfaces (API and web interface) of the database structure the data and their metadata. In this way the database tries to ensure for instance the submission of all relevant metadata belonging to a dataset. Another example is the automatically generated primary key, which ensures that every row in a table has got an unique identifier. For further information see also DatabaseRules.
- Plausibility and integration tests are applied to identify mistakes in the data, e.g. the geometrical correctness test which makes the manual testing with SQL-functions obsolete.
- When the number of users becomes large enough, a third level of data testing and rating based on user evaluations might be implemented.
Users can contribute to data quality for instance by following the DatabaseRules. For further reading and guidelines with regard to data management and publication see for instance: Open Knowledge Foundation, Open Data Foundation and Software Carpentry (e.g. here).
