Ruby port of TinySegmenter.js for tokenizing Japanese text. Uses a Naive Bayes model that has been trained using the RWCP corpus and optimized using L1-norm regularization. The resultant model is quite compact, yet has a 95% accuracy rate.
Required Ruby Version
>= 0
Authors
Peter Graham
Versions
- 0.0.6 October 26, 2015 (16 KB)
- 0.0.4 March 31, 2013 (16 KB)
- 0.0.2 August 27, 2012 (14 KB)
- 0.0.1 August 20, 2012 (11.5 KB)