NLTK Toolkit
This is a simple Python wrapper for the Natural Language Toolkit and scikit-learn. It was written for use in the classroom to play with text mining and topic modeling tools by examining an input file of "yaks" obtained from the YikYak social media application.
The nltktool script requires the following Python libraries:
- Python 2.7+ or 3.x
- numpy
- scipy
- nltk 3.1+
- sklearn
Viewing some basic information from input file:
$ nltktool -f yaks-stripped.txt stats
opening file: yaks-stripped.txt
total number of lines: 3698
total number of words: 56019
total number of unique non-stop
lexical variety: 9.68669239805
words: 5750
The '--help' option displays some basic usage instructions for the script and a list of supported functions
Please email Jed Dobson (james.e.dobson@dartmouth.edu) with any problems, fixes, or suggestions.
Web: http://www.cultcritlab.org
49dd75922b6e38932fec05290e2e9be4b83e16a6