NSLinguisticTagger for languages other than English

Hello all,

I'm currently checking out Swift's NSLinguisticTagger. For test purposes I used the code from appcoda Introduction to Natural Language Processing.

For the English language it works as expected and described in the tutorial. But when I use NSLinguisticTagger on languages other than English the Lemmatization, Parts of Speech and Named Entity Recognition produces no useful results. I can understand this for the Named Entity Recognition but for the first two options I thought at least a basic result should be possible.

For the English sentence the result is exactly as expected.

But for a German sentence only the language detection and the tokenization seems to work correct. For the Parts of Speech Tagging only "OtherWord" and for the Named Entity Recognition no result at all is returned.

Did anyone tried to use the Class in other languages than English or is it only seriously usable when working with English text. I couldn't find any Apple documentation explaining the language capabilities beside from a list of languages that should be supported.

Any comment pointing me to a solution is greatly appreciated.


NSLinguisticTagger is not part of the Swift open source project. It's part of Apple's NLP frameworks.

That said, you might want to take a look at the newer NL framework by Apple.