Difference between revisions of "Part of Speech Tagging"
From Cohen Courses
Jump to navigationJump to searchPastStudents (talk | contribs) |
PastStudents (talk | contribs) |
||
| Line 12: | Line 12: | ||
== Example Systems == | == Example Systems == | ||
* [http://www.markwatson.com/opensource/ FastTag] - open source implementation of Brill Tagger | * [http://www.markwatson.com/opensource/ FastTag] - open source implementation of Brill Tagger | ||
| + | * [http://nlp.stanford.edu/software/tagger.shtml Stanford Log-linear Part-of-Speech Tagger] | ||
* [http://opennlp.sourceforge.net/ OpenNLP Tagger] - based on maximum entropy | * [http://opennlp.sourceforge.net/ OpenNLP Tagger] - based on maximum entropy | ||
* [http://crftagger.sourceforge.net/ CRF Tagger] - based on conditional random fields | * [http://crftagger.sourceforge.net/ CRF Tagger] - based on conditional random fields | ||
Revision as of 19:12, 31 October 2010
Summary
Part of Speech Tagging (or POS Tagging for short) is a task in the field of computational linguistics which looks at marking a text corpus with the associated word categories known as parts of speech to words.
Common Approaches
Some common approaches to POS Tagging include the following:
- Hidden Markov Models based approaches, sometimes referred to as stochastic algorithms in older literature
- Dynamic Programming/Viterbi-like algorithms (DeRose & Church)
- Unsupervised approaches: Brill Tagger (Transformation-based learning), Constraint Grammar, Forward-Backward
Example Systems
- FastTag - open source implementation of Brill Tagger
- Stanford Log-linear Part-of-Speech Tagger
- OpenNLP Tagger - based on maximum entropy
- CRF Tagger - based on conditional random fields
- LingPipe - tool kit that contains models for POS tagging
