User:Ilienert/Using apertium-tagger

A basic guide to apertium-tagger

Once the tagger has been trained, it generates a .prob file. Once this has been generated, we can call apertium-tagger on it to disambiguate the tagged corpus.

For example, assume we are working with apertium-en-es. The simplest way to test out apertium-tagger would be something like the following:

echo "he sent" | /usr/local/bin/lt-proc /usr/local/share/apertium/apertium-en-es/en-es.automorf.bin | \ 
/usr/local/bin/apertium-tagger -g $2 /usr/local/share/apertium/apertium-en-es/en-es.prob

If you wish to see each lexical form corresponding to their occurring surface forms, with the tagger-chosen one printed first, use the "-f" or "--first" flags.

For more information, consult man apertium-tagger.