Add WordCloud

@@ -2,10 +2,15 @@ import os
 import csv
 import re
 
+import matplotlib.pyplot as plt
+import numpy as np
 from konlpy.tag import Okt
 from konlpy.tag import Komoran
+from PIL import Image
 
 from textrank import KeywordSummarizer
+from wordcloud import WordCloud, get_single_color_func
+
 
 okt = Okt()
 def Okt_tokenizer(sent):
@@ -15,9 +20,13 @@ def Okt_tokenizer(sent):
 komoran = Komoran()
 def komoran_tokenizer(sent):
     words = komoran.pos(sent, join=True)
-    words = [w for w in words if('/NNG' in w or '/NNP' in w)]
+    words = [w for w in words if('/NNG' in w or '/NNP' in w) and (len(w) > 5)]
+    
     return words
 
+def color_func(word, font_size, position,orientation,random_state=None, **kwargs):
+    return("hsl(0, 100%, {:d}%)".format(np.random.randint(40,65)))
+
 BASE_DIR = os.path.dirname(os.path.abspath(__file__))
 
 def GetKeywords():
@@ -33,9 +42,32 @@ def GetKeywords():
         verbose= False
     )
 
-    keywords = keyword_extractor.summarize(posts, topk=30)
-    print(keywords)
-    exit()
+    keywords = keyword_extractor.summarize(posts, topk=100)
     return keywords
 
-GetKeywords()
\ No newline at end of file
+def GetWordCloud():
+    keywords = GetKeywords()
+    texts = {}
+    for word in keywords:
+        texts[word[0][0:-4]] = word[1]
+
+    khu_mask = np.array(Image.open('./static/images/khu_lion.png'))
+
+    wordcloud = WordCloud(
+        font_path = './static/fonts/NanumGothicBold.ttf',
+        width = 800,
+        height = 600,
+        background_color="white",
+        mask = khu_mask,
+        color_func = color_func
+    )
+    wordcloud = wordcloud.generate_from_frequencies(texts)
+    array = wordcloud.to_array()
+
+    fig = plt.figure(figsize=(10,10))
+    plt.imshow(array, interpolation="bilinear")
+    plt.axis("off")
+    plt.show()
+    fig.savefig('./static/images/wordcloud.png')
+
+GetWordCloud()
\ No newline at end of file