Wikidata cleaned for errors and consistent data types
WiCleanData: Guaranteeing the Type Consistency of Wikidata by Taxonomy Refinement and Constraint Enforcement
Artificial Intelligence
Summary
Wikidata is a big collaborative database that has problems like errors and confusing categories. Manually fixing these issues is too hard because of its size. The authors created WiCleanData, an automated process that uses language models to tidy up the categories, simplify rules about data types, and remove conflicting facts. This results in a version of Wikidata without type mistakes, which anyone can explore and use for other projects.
What this means in practice
- •For knowledge graph engineers: Use WiCleanData’s consistent taxonomy and free-of-error facts to build more reliable semantic applications and query engines.
- •For data integration teams: Combine external datasets with Wikidata more safely by relying on WiCleanData’s validated and simpler type constraints.
Authors
Yiwen Peng, Marc Jeanmougin, Thomas Bonald
Abstract
Because of its collaborative nature, Wikidata suffers from errors, in- consistencies, and excessive complexity, such as redundant classes, ambiguity between instances and classes, wrong taxonomic paths, and type constraint violations. The manual curation of these issues is infeasible at scale. To address these challenges, we introduce WiCleanData, a refined version of Wikidata with a consistent tax- onomy and free from type constraint violations. Specifically, we have designed an automated pipeline that first cleans the taxonomy with language model assistance, then simplifies type constraints by hierarchical aggregation, and finally filters facts accordingly. The resulting knowledge graph, free from any type violation, is made publicly available via a Web interface, enabling easy exploration and downstream applications.