Part-of-speech tagging

Part-of-speech tagging is the task of assigning a part-of-speech tag to each word in a given text.

Software

• ACOPOST - a collection of taggers using maximum entropy, second order Markov, exemplar, and transformation-based models. See also this site. Free, open source license.
• Citar - uses trigram-based HMMs. Free, open source license.
• CRF++ - uses Conditional random fields. Free, open source license (dual: LGPL, New BSD). C++.
• CRFTagger - for English. Free, open source license. Java.
• GPoSTTL - Enhanced TBL tagger for English. Open source license.
• HunPos - uses trigram-based HMMs. Free, open source license. OCaml.
• Illinois LBJ POS Tagger - Uses averaged Perceptron based sequential model. Java API, Free, open source license.
• Memory-based tagger (MBT) - uses TiMBL. Free, open source license.
• Morče - Uses Averaged Perceptron based model. Free, open source license (GPL2).
• Sequor - Uses Averaged Perceptron based model. Free, open source license (New BSD). Haskell.
• Stanford Tagger - uses Maximum entropy models. Free, open source license. Java.
• SVMTool - uses Support vector machines. Free, open source license, but depends on non-Free/open source SVMlight.

See also

• POS Tagging (State of the art)

External links

• Wikipedia article on POS tagging