The GDELT Project

Global Similarity Graph: 26.7% Of Links Are Cross-Language

In its first week of operation, the new Global Similarity Graph has recorded 3.6M similarity connections between articles, of which 972K were between articles published in different languages! In all, 26.7% of the similarity links recorded thus far are cross-language, showcasing the tremendous power of this new dataset for allowing the tracing of stories across languages!

Learn More.