PyNLP is designed to perform simple NLP procedures (examples include tokenizing, removing stop words, frequency analysis,...)


Design Features:
	- Light weight
	- PyNLP is multi-core friendly. It will utilize all cores for its operations. 
	- For each module unittests are provided as part of the source code. Unit tests cover most of the functions and modules


Ideas for the future:
	- Combine Tokenize and Preprocess into one package

Help:
	- preprocess contains functions for tokenizing
	- dataTools helps you import your data


Please note: This project is still in very early stages. Please report bugs to faridani@berkeley.edu 



Siamak Faridani
UC Berkeley
July 2010
faridani@berkeley.edu