brown_wordlist.py

# Export wordlists of differently tagged word types from brown corpus
# using the natural language toolkit.

import io
import nltk

brown_tagged = nltk.corpus.brown.tagged_words(tagset='universal')
word_tag_fd = nltk.FreqDist(brown_tagged)

types = ['NOUN', 'VERB', 'DET', 'ADJ', 'ADP', 'CONJ', 'ADV', 'PRT']
for t in types:
    words = [wt[0] for (wt, _) in word_tag_fd.most_common() if wt[1] == t]
    with io.open("{0}.txt".format(t), 'w', encoding='utf8') as out:
        out.write('\n'.join(words))