wordcloud.py

#coding:utf-8
import csv
from janome.tokenizer import Tokenizer
from wordcloud import WordCloud
from collections import defaultdict
import datetime

exclude_words = [u'これ', u'こと', u'そう', u'それ', u'ところ', u'みたい', u'よう', u'さん',
    'http', 'https', 'www', 'to', 'htn', 'co', 'com', 'jp', 'ly', 'tw', 'RT', 'tinyurl', 'Bookmarking']

# 指定年の Tweet 一覧を取得 (Twitter からダウンロードした tweet.csv を使用)
def get_tweets(year):
    import csv
    with open('./tweets/tweets.csv', 'r', encoding='utf-8') as f:
        reader = csv.reader(f, delimiter=',')
        next(reader)
        for row in reader:
            if len(row) <= 5:
                continue

            dt = datetime.datetime.strptime(row[3], '%Y-%m-%d %H:%M:%S +0000')
            content = row[5]
            if dt.year == year:
                yield content

# 名詞だけ抽出
def get_words(texts):
    t = Tokenizer()
    words = []
    for text in texts:
        tokens = t.tokenize(text)
        for token in tokens:
            pos = token.part_of_speech.split(',')[0]
            if pos == '名詞':
                words.append(token.base_form.strip())
    return words

for year in range(2007, 2018):
    words = get_words(get_tweets(year))
    text = ' '.join(words)
    wordcloud = WordCloud(background_color='white', font_path='NotoSansCJKjp-Regular.otf',
        stopwords=exclude_words, collocations=False, width=800, height=800).generate(text)
    wordcloud.to_file(f'./{year}.png')