Multilanguage support without some languages?

I was just browsing github and found an old commit that stated that there is language support for chinese. And on sourceforge there are some older models (2016-10) that include arabic and chinese.

I was wondering why these languages not included in the newest “multilingual” dbpedia-spotlight-1.1.jar.jar file? And are you planning on releasing a file with these languages included?

1 Like

Hi @mingaflo,

Thanks for the question. The main idea is to extend DBpedia Spotlight to all available languages. The model creation works but we found some issues testing the model with the DBpedia Spotlight service. The problem is related to the stemming process; we need to find stemmer algorithms to support the additional languages (e.g., Chinese). I hope this information helps. Thanks again, have a great day.

Maybe a GSoC project?

1 Like