This toolkit can be used to analyse a set of word pairs (base-form + modified form) provided. Analysis is based on surface form orthography.
Currently it can only: (a) learn and predict affixation, null derivation, clipping and periphrasis, (b) display derived rule sets as networks (c) extract and draw categorisation networks
Hopefully in future it will be able to do more!
This code is still work in progress - until downloads are tagged and made available, code can change without notice.