Update app.py

app.py

@@ -9,7 +9,7 @@ client = InferenceClient ('Qwen/Qwen2.5-72B-Instruct')
 
 # Step 1
 with open("Mental health chatbot text.txt", "r", encoding="utf-8") as f:
-    skincare_text = f.read()
+    mental_health_text = f.read()
 
 # Step 2: Preprocess text into sentence chunks
 def preprocess_text(text):
@@ -39,7 +39,7 @@ def preprocess_text(text):
     print(f"Sample chunks: {final_chunks[:3]}")
     return final_chunks
 
-cleaned_chunks = preprocess_text(skincare_text)
+cleaned_chunks = preprocess_text(mental_health_text)
 
 # Step 3: Convert chunks into embeddings
 model = SentenceTransformer('all-MiniLM-L6-v2')
@@ -68,22 +68,23 @@ def get_top_chunks(query, chunk_embeddings, text_chunks, top_k=3):
     return results, top_scores[:len(results)]
 
 # Step 5: Relevance checker
-def is_skincare_related(query):
-    skincare_keywords = [
-        'skin', 'skincare', 'acne', 'wrinkles', 'moisturizer', 'cleanser', 
-        'sunscreen', 'serum', 'retinol', 'vitamin', 'dry', 'oily', 'sensitive',
-        'aging', 'pores', 'blackheads', 'routine', 'face', 'facial', 'beauty',
-        'dermatology', 'cosmetic', 'cream', 'lotion', 'toner', 'exfoliate',
-        'hydration', 'anti-aging', 'blemish', 'spot', 'dark circles'
+def is_mental_health_related(query):
+    mental_health_keywords = [
+        'anxiety', 'depression', 'stress', 'mental health', 'therapy', 'counseling',
+        'mood', 'emotions', 'feelings', 'wellbeing', 'self-care', 'mindfulness',
+        'meditation', 'coping', 'support', 'psychology', 'psychiatry', 'bipolar',
+        'trauma', 'PTSD', 'panic', 'worry', 'sad', 'happy', 'angry', 'fear',
+        'self-esteem', 'confidence', 'resilience', 'healing', 'recovery',
+        'mental', 'emotional', 'psychological', 'behavioral', 'cognitive'
     ]
     
     query_lower = query.lower()
-    return any(keyword in query_lower for keyword in skincare_keywords)
+    return any(keyword in query_lower for keyword in mental_health_keywords)
 
 queries = [
-    "Consistent skincare routine",
-    "Applying sunscreen daily", 
-    "Choosing products that match your skin type"
+    "Managing daily stress and anxiety",
+    "Building healthy coping mechanisms", 
+    "Practicing mindfulness and self-care"
 ]
 
 for q in queries:
@@ -96,7 +97,7 @@ def respond(message, history):
     top_results = get_top_chunks(message, chunk_embeddings, cleaned_chunks)
     print(top_results)
     
-    messages = [{"role": "system", "content": f"You are a friendly chatbot. You give people advice about skincare. Base your response on the following information: {top_results}"}]
+    messages = [{"role": "system", "content": f"You are a friendly chatbot. You give people advice about mental-health. Base your response on the following information: {top_results}"}]
     
     if history:
         messages.extend(history)