corpus/           - Various versions of the corpus. Only one of them is actually needed (the default). 
data/             - all the results get written there
sampleInputFile/  - how an input file for textchunker.py should look like
PruneChunker.py   - most of the code is in there.. (not runnable)
extract.py        - used to extract all NP rules from a given corpus
chunk.py          - evaluate set of chunking rules on a given corpus
prune.py          - prune a ruleset based on a given corpus
textchunker.py    - chunk given tagged text for using given ruleset