TnT - Thorsten Brants's part-of-speech tagger

TnT is a part-of-speech tagger (POS-tagger) which can be used to prepare corpora for search tools that presume POS tagging (e.g. Gsearch, tgrep, etc.). It comes pretrained to tag English and German newspaper text but can be trained with any other corpus. The Unix and Windows version can be found on AFS at:

    /afs/ir/data/linguistic-data/lib/tnt

Tip - Jeanette Pettibone has provided us with her presentation on part-of-speech taggers.