The pyzeta scripts are a Python implementation of Craig's Zeta score for contrastive text analysis.
This script implements a relatively simple measure for distinctive features in two groups of texts. It allows you to find out which words are characteristic of one group of texts when compared to another group of texts.
The underlying measure of distinctiveness or keyness has been proposed by John Burrows under the name of Zeta
hence the name of this Python package.
- The
sampledata
folder contains some examples of what input pyzeta needs and what output it produces - The
howto.md
file contains a brief tutorial of sorts for running analyses with pyzeta.
- Requirements: Python 3 with pandas, numpy, sklearn, pygal, treetaggerwrapper (and TreeTagger)