Sequence Tagging powered by the Averaged Perceptron.
Part of Speech Tagging
A Part of Speech tagger using the Average Perceptron.
Based on the tagger from here
This uses the following features:
- The Suffix (last 3 characters) of the current word (unnormalized).
- The Prefix (first character) of the current word (unnormalized).
- The current word.
- The previous Part of Speech tag and the current word.
- The Previous Part of Speech tag.
- The Part of Speech tag from the word before last.
- Both of the previous Part of Speech tags.
- The previous word.
- The previous word suffix.
- The word from 2 steps back.
- The next word.
- The next word suffix.
- The word after next.
- A Bias
Includes the following Pretrained models.
- POS Tagger, Trained on the CoNLL 2000 Chunking data
- Chunker, Trained on the CoNLL 2000 Chunking data
- Slot filler, Trained on ATIS data
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
sequence_tagging-0.1.6.tar.gz (2.9 MB view hashes)