워드 클라우드로 소설을 보는 단어.

from janome.tokenizer import Tokenizer
import zipfile
import os.path, urllib.request as request
from wordcloud import WordCloud
import matplotlib.pyplot as plt
%matplotlib inline


file = withopen('rashomon.txt', 'r') as f:
    bindata = f.read()
textdata = bindata.decode('shift_jis')


t=Tokenizer()
tokens=t.tokenize(textdata)

words = ""
for token in tokens:
    if token.part_of_speech.split(',')[0] in ['名詞', '動詞', '副詞']:
        words = words + " " + token.base_form


fpath="NotoSansCJKjp-hinted/NotoSansCJKjp-Black.otf"
wordcloud = WordCloud(background_color="white",width=800,height=500,font_path=fpath).generate(words)


plt.figure(figsize=(30,24))
plt.imshow(wordcloud)
plt.axis("off")
plt.show()