Workshop on Data Quality Management in Wikidata
3 likes653 views
The document discusses the importance of data quality in open data, outlining its dimensions, metrics, assessment methods, and improvement strategies. It highlights challenges such as inaccuracies and inconsistencies in datasets, emphasizing the significance of accessibility, completeness, and interlinking in data quality. Various tools and crowdsourcing methods for assessing and improving linked data quality are also presented.
Workshop on Data Quality Management in Wikidata
- 2. 2
- 4. 4 Garbage Garbage Biggest Challenge: Poor Data Quality *http://www.ibmbigdatahub.com/infographic/four-vs-big-data @amrapaliz
- 6. Systematic Literature Review 30 core articles Conference - 21 Journal - 8 Masters Thesis - 1 18 Dimensions 69 Metrics @amrapaliz
- 7. LDQ Dimensions & Metrics Quality assessment for linked data: A survey. A Zaveri, A Rula, A Maurino, R Pietrobon, J Lehmann, S Auer. Semantic Web 7 (1), 63-93 300+ citations @amrapaliz
- 8. LDQ Dimensions & Metrics •Data Quality: commonly conceived as a multi- dimensional construct with a popular definition ‘fitness for use’*. •Dimension: characteristics of a dataset. •Metric: or indicator is a procedure for measuring an information quality dimension. *Juran et al., The Quality Control Handbook, 1974 @amrapaliz
- 9. LDQ Assessment Goal Fix data quality issues in given sets of (semantic) data Such quality issues may • be in source datasets (e.g., inaccurate or wrong data items, outdated data items) • result from imperfections of a data integration process (e.g., data items that have been incorrectly linked with each other) • reveal themselves only after the data integration (e.g., duplicates, inconsistencies) Data cleaning may be relevant both, for original datasets before combining/ integrating and for datasets resulting from an integration. Source: http://www.ida.liu.se/research/semanticweb/events/SemDataMgmtTutorial-Part7- Cleaning.pdf @amrapaliz
- 11. LDQ Dimensions - Accessibility dimensions & metrics • Availability - extent to which data (or some portion of it) is present, obtainable and ready for use • accessibility of the SPARQL endpoint and the server • dereferenceability of the URI • Interlinking - degree to which entities that represent the same concept are linked to each other, be it within or between two or more data sources • detection of the existence and usage of external URIs • detection of all local in-links or back-links: all triples from a dataset that have the resource’s URI as the object @amrapaliz
- 12. LDQ Dimensions - Intrinsic dimensions & metrics • Syntactic Validity - degree to which an RDF document conforms to the specification of the serialization format • detecting syntax errors using (i) validators, (ii) via crowdsourcing • by (i) use of explicit definition of the allowed values for a datatype, (ii) syntactic rules (type of characters allowed and/ or the pattern of literal values) • @amrapaliz
- 13. LDQ Dimensions - Intrinsic dimensions & metrics • Completeness • Schema - ontology completeness • Property - missing values for a specific property • Population - % of all real-world objects of a particular type • Interlinking - degree to which instances in the dataset are interlinked @amrapaliz
- 15. RDFUnit: RDF Unit-Testing Suite http://aksw.org/Projects/RDFUnit.html Syntactic Semantic Consiste @amrapaliz
- 16. 16 Crowdsourcing Linked Data Quality Assessment Crowdsourcing linked data quality assessment M Acosta, A Zaveri, E Simperl, D Kontokostas, S Auer, J Lehmann ISWC 2013 @amrapaliz
- 17. Luzzu: QA for LOD http://eis-bonn.github.io/Luzzu/index.html 2 Asses 3 Clean 4 Store 5 Rank 1 Metric @amrapaliz
- 18. LDQ Assessment Tools — LODLaundromat http://lodlaundromat.org/ @amrapaliz
- 19. LDQ Beyond Data — Mapping Quality Dimou et al. Assessing and Refining Mappings to RDF to Improve Dataset Quality. ISWC 2015. https://github.com/RMLio/RML-Validator @amrapaliz
- 22. Data Quality Improvement • Root cause analysis • Iterative process • Ensure high data and metadata quality @amrapaliz
- 25. 25 *400 AI papers 6% code 30% test data 54% pseudo code *http://science.sciencemag.org/content/359/6377/725 @amrapaliz
- 26. 26 Lambin et al. Radiother Oncol. 2013. 109(1):159-64. doi: 10.1016/j.radonc.2013.07.007 @amrapaliz
- 30. Improving the FAIRness of digital resources will increase their quality and their potential and ease for reuse. 30 @amrapaliz