MVP almost done

twitterPredictor/twitterCollect/dataframe.py

@@ -37,8 +37,7 @@ def convert_2_dataframe(data):
             hash_list.append("#"+hash.get("text"))
 
         hashtags.append(hash_list)
-
-        print(tweet.retweet_count)
+        
         retweets.append(tweet.retweet_count)
         likes.append(tweet.favorite_count)
 

twitterPredictor/twitterCollect/opinion.py

+from collect_candidate_tweet_activity import *
+from dataframe import *
+from textblob import *
+
+def categorize_tweets(data,neutral_line):
+    pos_tweets = []
+    neu_tweets = []
+    neg_tweets = []
+
+    for item in data["text"]:
+        try:
+            blob = TextBlob(item)
+            blob = blob.translate(to='en')
+        except:
+            blob = TextBlob(item)
+
+        polarity = blob.sentiment.polarity
+        print(blob)
+        print(polarity)
+        if polarity<=neutral_line and polarity >=-neutral_line:
+            neu_tweets.append(item)
+        elif polarity > neutral_line:
+            pos_tweets.append(item)
+        else:
+            neg_tweets.append(item)
+
+    return pos_tweets,neu_tweets,neg_tweets
+
+tweets = get_replies_to_candidate("EmmanuelMacron")
+data = convert_2_dataframe(tweets)
+
+pos_tweets,neu_tweets,neg_tweets = categorize_tweets(data,0.1)
+
+print("Percentage of positive tweets: {}%".format(len(pos_tweets)*100/len(data['text'])))
+print("Percentage of neutral tweets: {}%".format(len(neu_tweets)*100/len(data['text'])))
+print("Percentage de negative tweets: {}%".format(len(neg_tweets)*100/len(data['text'])))