Language detection module is ngram.erl.  Based on ruby version found
on github.com, http://github.com/feedbackmine/language_detector

See the wikipedia article for more information about this method.
http://en.wikipedia.org/wiki/Ngram

To run tests use eunit  eunit:test('ngram')

To build type `make`

The beam files will be put in the ebin directory.  The training data
is in the data directory.  This data is taken from wikipedia.  See
source code for missing languages.