Skip to content

Latest commit

 

History

History
41 lines (26 loc) · 1.88 KB

README.md

File metadata and controls

41 lines (26 loc) · 1.88 KB
Raw

Wikipedia word frequency generator

This script processes wikipedia article dumps from https://dumps.wikimedia.org/enwiki/ and gathers the word frequency distribution data. The script uses wikiextractor to fetch raw text, then strips punctuation marks and normalizes unicode dashes and apostrophes. The script then disregards words that have a digit in them, and only takes words that were used in at least 3 different articles.

The script was inspired by this article which unfortunately provided very inaccurate data with punctuation marks and other sorts of inaccuracies.

Usage

The script needs Python 3. On macOS, there is a known bug with Python 3.8, so you will need to use Python 3.7 or lower.

Install requirements:

pip install -r requirements.txt

Download the current Wikipedia dumps for the desired language:

WIKI=enwiki
wget -np -r --accept-regex \
  "https:\/\/dumps\.wikimedia\.org\/${WIKI}\/latest\/${WIKI}-latest-pages-articles[0-9]*\.xml.bz2" \
  https://dumps.wikimedia.org/${WIKI}/latest/

Note that for enwiki (as of April 2023) this will require about 19 Gb of free space.

Parse dumps and save results:

python ./gather_wordfreq.py dumps.wikimedia.org/${WIKI}/latest/*.bz2 > wordfreq.txt

Pre-generated word frequency data

The word frequency data for English, Spanish, French, German, Italian, Portuguese, Dutch, Arabic, Polish, Egyptian, Japanese, Russian, Cebuano, Swedish, Ukrainian, Vietnamese, Chinese, Waray, Afrikaans & Swahili are provided at results.

English results:

  • Total unique words appearing at least in 3 articles: 2747823
  • Top 20 most popular words: the, of, in, and, a, to, was, is, on, for, as, with, by, he, that, at, from, his, it, an.