Birdcloud.py

# Tweet to Twitter Birdcloud (Wordcloud)  
  
import pandas as pd  
import re  
import os  
os.chdir("C:/Users/Dome/Desktop/nu/Tweets/")  
  
party= "linksfraktion"  
df=pd.read_json(party+".json")  
  
from nltk.tokenize import TweetTokenizer  
from stop_words import get_stop_words  
sw = get_stop_words('de')  
tt = TweetTokenizer()  
sw.extend(("macht","schen","schon","Kapitel","setzt","weitere","lehnt","viel",  
           "stellen","gehen","chen","geht","gilt","lehnen", "viele","gibt",  
           "darf","halten","dürfen","neben","gehört","vielen","jedoch",  
           "braucht", "nehmen","rung","seit","sollten","deren","etwa",  
           "beim","außerdem","stehen", "sitzen", "mehr","sollen","müssen",  
           "sowie","deshalb","daher","dafür","dabei","brauchen","zudem",  
           "setzen","besser","neue","neu","neuen","immer","gute"))  
  
df['tokens'] = df['text'].apply(tt.tokenize)   
  
# complicated expression:   
# all items not in stopwords, or matching @/#  
df['clean'] = df['tokens'].apply(lambda x: [item for item in x if   
  item.isalpha() and item.lower() not in sw or re.compile(r"@|#").match(item)   
          and item.lower() not in sw])   
###### clean df! ######  
search_values = ["migr","flücht","auslä","asyl","flucht","immigr","refugee",  
                 "geflüchte", "ausland", "zuwander", "zugewandert"]  
mig=df[df.text.str.contains('|'.join(search_values ))]  
  
from wordcloud import WordCloud, STOPWORDS, ImageColorGenerator   
from PIL import Image  
import numpy as np  
import matplotlib.pyplot as plt  
  
# list flattening, as here we have a list in a list  
flattened  = [val for sublist in list(mig.clean) for val in sublist]  
  
words = ' '.join(flattened) # list to text  
  
# define twitter bird mask, must have white background  
mask = np.array(Image.open("C:/Users/Dome/Desktop/nu/Tweets/Wordclouds/Twitter_bird.png"))  
image_colors = ImageColorGenerator(mask)  
# function takes in your text and your mask and generates a wordcloud.   
def birdcloud(words, mask):  
    word_cloud = WordCloud(width = 512, height = 512,max_words=200,  
                           background_color='white', stopwords=STOPWORDS,  
                           mask=mask).generate(words)  
    plt.figure(figsize=(10,8),facecolor = 'white', edgecolor='blue')  
    plt.imshow(word_cloud.recolor(color_func=image_colors),  
               interpolation="bilinear")  
    plt.axis('off')  
    plt.tight_layout(pad=0)  
    plt.savefig("C:/Users/Dome/Desktop/nu/Tweets/Wordclouds/"+  
                party+".png", format="png")  
    plt.show()  
      
# generate your birdcloud  
birdcloud(words, mask)