remove word less than 4

preprocessor.py

@@ -39,10 +39,10 @@ def preprocess(data):
 
     def clean_message(message):
         # Remove messages containing '<Media ...>'
-        if '<Media' in message:
+        if '<Médias omis>' in message:
             return ''
-        # Remove words with less than 3 characters
-        words = [word for word in message.split() if len(word) >= 3]
+        # Remove words with less than 4 characters
+        words = [word for word in message.split() if len(word) >= 4]
         # Remove stopwords
         words = [word for word in words if word.lower() not in combined_stop_words]
         return ' '.join(words)