wikidata_translation

Wikidata is a multilingual, collaborative and structured knowledge base. Humans actively edit, update and remove the elements of this database over more than two hundreds languages. Some tools exist to visualize those revisions (WD-Prop). Nevertheless, it could be interesting to analyze the information flow hidden in the translation process of the Wikidata label properties. Using frequent pattern mining methods and visualizations, this project aims to show how to detect correlations or translation patterns between different languages.

Notebooks:

• Get Data

• Frequent Pattern Minning

• Network Analysis

Final repport:

• link pdf