google

app.py

@@ -34,6 +34,13 @@ model_choices = {
 
 model_cache = {}
 
+def emphasize_keywords(text, keywords, repeat=3):
+    for kw in keywords:
+        pattern = r'\b' + re.escape(kw) + r'\b'
+        text = re.sub(pattern, (kw + ' ') * repeat, text, flags=re.IGNORECASE)
+    return text
+
+
 # Clean text: remove special characters and stop words
 def clean_text(input_text):
     cleaned = re.sub(r"[^A-Za-z0-9\s]", " ", input_text)
@@ -41,6 +48,12 @@ def clean_text(input_text):
 
     words = cleaned.split()
     words = [word for word in words if word.lower() not in stop_words]
+
+        # Example keyword list
+    keywords = ["blazer", "shirt", "trouser", "saree", "tie", "suit"]
+
+    words = emphasize_keywords(words, keywords)
+
     return " ".join(words).strip()
 
 # Load model and tokenizer