Code
- Unsupervised POS Induction
[C++ source]
[amd64 static binary]
(Has enhancements over the conference version. See README.)
- Unsupervised Word Morphological Segmentation for Arabic
[Python source]
- Above Segmenter for Large Data Sets
[C++ source]
(Performs maximal marginal decoding and has heuristics for handling large corpus. See README.)