improve german word selection

hitler_texts/parse.py

@@ -6,7 +6,15 @@ with open('/home/josua/deu_mixed-typical_2011_1M/deu_mixed-typical_2011_1M-words
     for line in lines:
         #print(line.split(sep="\t"))
         index, word, count = line.split(sep="\t")
-        if int(index) > 100:
+        if int(index) > 100 and int(count) > 5:
+            german_words.append(word.lower())
+
+with open('/home/josua/deu_mixed-typical_2011_1M/deu_news_1995_1M-words.txt') as f:
+    lines = f.readlines()
+    for line in lines:
+        #print(line.split(sep="\t"))
+        index, word, count = line.split(sep="\t")
+        if int(index) > 100 and int(count) > 5:# only words that are used more than 5 times
             german_words.append(word.lower())
 
 
@@ -36,11 +44,14 @@ with open(f'mein_kampf') as f:
     for line in lines:
         hitler_words.extend(get_words_from_line(line))
 
+german_words = set(german_words)
 hitler_words = set(hitler_words) #unique
 #filter_words = hitler_words.intersection(set(german_words))
 
 only_hitler_words = list(hitler_words.difference(german_words))
 
+print(only_hitler_words)
+
 with open("german_words", "w") as f:
     for word in german_words:
         word += "\n"