Some quick and dirty code for calculating Burrows's Delta measure of textual difference. Solutions like https://files.nyu.edu/dh3/public/TheDeltaSpreadsheets.html and https://sites.google.com/site/computationalstylistics/ already exist but I did not find any Python code.

Needs NLTK, matplotlib, numpy

Run on command line with "python delta.py [author corpuses] -t target text" Additional optional switches include -c to set culling of words from main corpus (70, for example, would cull if 70 of a specific word is in the total corpus by virtue of its abundance in a single author corpus) -f to set number of features to use (default is 500) -p [true/false] will plot the final Z scores if true -o [filename] will print output to a file (otherwise printed to stdout)