On the basis of the available corpora a quite exact classification can be accomplished by unknown text.
Identifies the most likely language of an unknown text.
Estimates and infers models present in a set of documents using the Topic Modeller of Mallet.