I use this library to detect the language of Tweets. It's not perfect, but does pretty well given it only has 140 characters to work with.
My Spanish users complain that quite often Portuguese tweets get identified as Spanish Tweets. One of them said that the letter ç only belongs to Portuguese. ã and ê may also be used to further differentiate.
I use this library to detect the language of Tweets. It's not perfect, but does pretty well given it only has 140 characters to work with.
My Spanish users complain that quite often Portuguese tweets get identified as Spanish Tweets. One of them said that the letter ç only belongs to Portuguese. ã and ê may also be used to further differentiate.