Unlike the rest of GDELT, which live translates from 65 languages, the neural Global Entity Graph (GEG) processes only English language news coverage, a legacy of GDELT being an early alpha adopter of the underlying Cloud Natural Language API when it first launched with only English support. Today, the API supports 11 languages, including Chinese Simplified, Chinese Traditional, English, French, German, Italian, Japanese, Korean, Portuguese (Brazilian & Continental), Russian and Spanish. Early next week GDELT will begin expanding its daily sample of articles processed through the GEG to encompass coverage from all 11 languages, allowing powerful explorations of multilingual neural language analysis.
As a first step, the neural GEG BigQuery table "gdelt-bq:gdeltv2.geg_gcnlapi" has been upgraded to include a new "lang" field that contains the Google-assigned language code for the document.