voc.txt was generated from a dump of the Hindi wikipedia by the script
wikipedia-most-common-words like so:

WikiExtractor.py hiwiki-latest-pages-articles.xml.bz2
cat text/*/* | scripts/wikipedia-most-common-words 10 devanagari > voc.txt

The dump used was dated May 3rd 2018.

output.txt was generated from voc.txt by running it through the stemmer:

stemwords -l hindi -c UTF_8 -i hindi/voc.txt -o hindi/output.txt

Wikipedia is licensed as: https://creativecommons.org/licenses/by-sa/3.0/
