|
# 국민 청원 접속 및 청원 목록 url 수집
url = []
for i in range(1,21):
html = urlopen("https://www1.president.go.kr/petitions/best?page={}".format(i))
soup = BeautifulSoup(html,'html.parser')
for j in soup.findAll('div',{'class':'bl_body'}):
for k in j.findAll('div',{'class':'bl_subject'}):
a = k.find('a')['href']
if bool(re.match('/[a-z].*/[0-9].*\?navigation=best',a)):
url.append(a)
# 청원 제목 수집
title = []
for i in url:
html = urlopen('https://www1.president.go.kr/'+i)
soup = BeautifulSoup(html, "html.parser")
for i in soup.findAll('h3',{'class':'petitionsView_title'}):
title.append(i.text)
# 문장 정제작업
txt = ''
for i in title :
txt = txt + ' ' + re.sub('[<|>|(|)|!|\,|/|\.]',' ',i)
# 단어별 빈도수 체크 / 불용어 처리
from konlpy.tag import Hannanum
hannanum = Hannanum()
text_list = hannanum.nouns(txt)
stopword = ['및','등']
ko = [i for i in text_list if i not in stopword]
ko = nltk.Text(ko)
data = ko.vocab()
# wordcloud 만들기
wordcloud = WordCloud(font_path='C:\windows/fonts/malgun.ttf',
background_color = 'white',
width = 1000, height = 800).generate_from_frequencies(dict(data))
plt.figure(figsize=(20,20))
plt.imshow(wordcloud)
plt.axis("off")
plt.show()
|
