- Fixes and improvements to the k-means clustering based reduction methods
- Added the hierarchical clustering based reduction method
- Support for "humanizing" final rule-bases
- Performance improvements
- Increased code readability
- Many small fixes / enhancements