2019-05-21 20:43:57 +00:00
|
|
|
# http://www.ulliwaltinger.de/sentiment/
|
|
|
|
|
# https://github.com/solariz/german_stopwords
|
|
|
|
|
#!/usr/bin/env python
|
2019-05-27 19:17:52 +00:00
|
|
|
# https://github.com/markuskiller/textblob-de
|
2019-05-21 20:43:57 +00:00
|
|
|
# -*- coding: utf-8 -*-
|
|
|
|
|
import nltk
|
|
|
|
|
import copy
|
|
|
|
|
import encodings
|
|
|
|
|
import csv
|
|
|
|
|
from siteobj import *
|
|
|
|
|
from nltk.corpus import treebank
|
2019-05-27 19:17:52 +00:00
|
|
|
from textblob_de import TextBlobDE as TextBlob
|
2019-05-21 20:43:57 +00:00
|
|
|
|
|
|
|
|
|
2019-05-27 19:17:52 +00:00
|
|
|
def get_sentiment(url):
|
|
|
|
|
NewsText = obj.read_article(url)
|
2019-05-21 20:43:57 +00:00
|
|
|
|
2019-05-27 19:17:52 +00:00
|
|
|
newText = ""
|
|
|
|
|
for text in NewsText:
|
|
|
|
|
newText += text
|
2019-05-21 20:43:57 +00:00
|
|
|
|
2019-05-27 19:17:52 +00:00
|
|
|
newText = TextBlob(newText)
|
2019-05-21 20:43:57 +00:00
|
|
|
|
2019-05-27 19:17:52 +00:00
|
|
|
sent = newText.sentiment[0]
|
|
|
|
|
if sent < 0:
|
|
|
|
|
good = "shit"
|
|
|
|
|
else:
|
|
|
|
|
good = "nice"
|
|
|
|
|
print(good, newText.sentiment,"\n", link.split("/")[-1], "\n")
|
|
|
|
|
return good
|
2019-05-21 20:43:57 +00:00
|
|
|
|
2019-05-27 19:17:52 +00:00
|
|
|
obj = Golem()
|
|
|
|
|
news, links = obj.get_news()
|
2019-05-21 20:43:57 +00:00
|
|
|
|
2019-05-27 19:17:52 +00:00
|
|
|
for link in links:
|
|
|
|
|
get_sentiment(link)
|