A research prepared by scientists from our Department was presented at Wiki Workshop 2025. The study delivers a comprehensive, topic-based analysis of Wikipedia articles across 55 language editions and employs an original approach that combines a citation index with a synthetic article-quality measure. A citation index was constructed by analysing 6.6 billion links between Wikipedia pages, enabling the identification of the most influential articles in each language edition.
The articles were grouped into 18 thematic categories using the open semantic knowledge base Wikidata, which also made it possible to pinpoint the most-cited pages within each topic. Every one of the 47 million articles was evaluated for quality with a synthetic measure that allows direct comparison even across language editions that apply different quality standards.
Presentation recording:
The results clearly reveal differences among language editions of Wikipedia across various topics. This work paves the way for further research on the quality of multilingual Wikipedia content and can inform targeted community efforts to improve articles in less-developed language editions.
The paper, “Utilizing citation index and synthetic quality measure to compare Wikipedia languages across various topics” is available open access. Authors of the work: Dr. Włodzimierz Lewoniewski, Prof. Krzysztof Węcel, Prof. Witold Abramowicz. Two associated open datasets have also been released: citation-index scores for all analysed Wikipedia articles on Hugging Face and article-quality assessments on Kaggle.
Wiki Workshop is an annual international conference organised by academic and expert communities studying Wikipedia and other Wikimedia Foundation projects. It seeks to foster the exchange of knowledge, experience, and research findings that support Wikipedia’s continued development and the improvement of its content. The 2025 edition was held online on 21–22 May 2025. Further details are available on the Wiki Workshop website: wikiworkshop.org.