optimaize / language-detector

Language Detection Library for Java
Apache License 2.0
567 stars 165 forks source link

"dallas" is recognised as Spanish Language #101

Open skilledmonster opened 4 years ago

skilledmonster commented 4 years ago

When I load English and Spanish languages, "dallas" which is a city name is recognized as Spanish text.

Is there an override setting for these type of inputs?

james-s-w-clark commented 4 years ago

@skilledmonster that's a very short string; Spanish just gets more "n-gram points" than English for it.

What do you mean by "these type of inputs"? Single word inputs?