Multilingual Support

Rosoka Software supports 230+ languages in a single application and can extract information from various sources, documents, or feeds. The source data can be in multiple languages, even when those languages are within an individual document. Rosoka Software also provides tokenization, language identification and English glossing for all the languages it supports.

Example

If a Korean document contains 블라디미르 푸틴, Rosoka Software will provide a transliteration of Vladimir Putin. Or if there is a name like 溫家寶, Rosoka Software will provide a Rosoka Software Entity Translation of Wen Jia-Bao. There are cases when you don’t want a transliteration, but actual translations of the entity. With something like, “國務院”, it makes more sense to see Department of State rather than only a transliteration of guo wu yuan.