Update analyzer.py

analyzer.py

@@ -6,16 +6,19 @@ import joblib
 from sklearn.feature_extraction.text import CountVectorizer
 from sklearn.naive_bayes import MultinomialNB
 from sklearn.pipeline import Pipeline
+from nltk.stem import PorterStemmer
 
 app = Flask(__name__)
 
 # Function to preprocess text data
 def clean_text(texts):
     cleaned_texts = []
+    ps = PorterStemmer()
     for text in texts:
         text = text.lower()
         text = re.sub(r'[^\w\s]', ' ', text)
-        cleaned_texts.append(text)
+        words = [ps.stem(word) for word in text.split()]
+        cleaned_texts.append(' '.join(words))
     return cleaned_texts
 
 # Function to determine sentiment label based on probability