PhD Proposal Targets Language Gaps in Knowledge Graphs
A new PhD proposal is addressing the gap in Open Access Data (OAD) between languages that have plenty of resources and those that don't, with a focus on Linked Open Data knowledge graphs (LOD KGs). It points out key aspects that influence language distribution, such as the number of Wikipedia articles available in different languages and the existence of language-tagged entities within LOD KGs. The research looks into three major multilingual LOD KGs: DBpedia, BabelNet, and Wikidata. The goal is to investigate how to choose candidates for cross-lingual transfer in multilingual KG completion, ultimately aiming to improve representation for underrepresented language communities. This proposal has been submitted to arXiv (ID: 2605.05931).
Key facts
- The proposal targets low-resource languages in Open Access Data.
- It analyzes language distribution in DBpedia, BabelNet, and Wikidata.
- Key variables include Wikipedia article counts and language-tagged entities.
- The research plans to study cross-lingual transfer for KG completion.
- The proposal is published on arXiv with ID 2605.05931.
- It aims to address the digital divide in global digital transformation.
- The study focuses on Linked Open Data knowledge graphs.
- Linguistic strategies are proposed for multilingual KG completion.
Entities
Institutions
- arXiv
- DBpedia
- BabelNet
- Wikidata