Feedback

ML/app.py

@@ -0,0 +1,24 @@
+from flask import Flask, request, jsonify, render_template
+from chat import chatbot
+
+app = Flask(__name__)
+
+@app.route("/")
+def start():
+    return render_template('home.html')
+
+@app.route("/chat", methods=['GET'])
+def chat_page():
+    return render_template('chat.html')
+
+@app.route("/chat", methods=["POST"])
+def chat():
+    data = request.get_json()
+    user_input = data.get("messageText")  
+    response = chatbot(user_input) 
+
+    return jsonify({"answer": response})  
+
+
+if __name__ == '__main__':
+    app.run(debug=True)

ML/chat.py

@@ -0,0 +1,88 @@
+import random
+import requests
+
+
+HUGGINGFACE_API_KEY = "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"
+
+
+EMOTION_ANALYSIS_MODEL = "j-hartmann/emotion-english-distilroberta-base"
+
+API_URL = "https://api-inference.huggingface.co/models"
+
+def analyze_emotion(text):
+    """Analyze nuanced emotions using Hugging Face emotion detection model."""
+    headers = {"Authorization": f"Bearer {HUGGINGFACE_API_KEY}"}
+    payload = {"inputs": text}
+
+    response = requests.post(f"{API_URL}/{EMOTION_ANALYSIS_MODEL}", headers=headers, json=payload)
+
+    if response.status_code == 200:
+        response_json = response.json()
+        print("Emotion Analysis Response:", response_json)
+
+        # Get the emotion with the highest score
+        emotions = response_json[0]
+        primary_emotion = max(emotions, key=lambda x: x['score'])['label'].lower()
+
+        return primary_emotion
+    else:
+        print("Emotion Analysis Failed with status code:", response.status_code)  
+        print("Response Content:", response.content)  
+        return "neutral"
+
+def generate_reframing_message(emotion):
+    """Generate a relevant message based on the detected emotion."""
+    reframing_messages = {
+        "anger": [
+            "Take a moment to reflect on what triggered this anger. Consider if there’s a constructive way to address it.",
+            "Anger can be challenging. It might help to channel this energy into a positive activity, like exercise or a creative task."
+        ],
+        "joy": [
+            "That's wonderful! Remember to savor this joyful moment.",
+            "It’s great to hear you’re feeling positive! Enjoy the happiness and think of ways to spread it around."
+        ],
+        "fear": [
+            "Fear can be overwhelming. You might consider listing out your concerns and thinking of small steps to address them.",
+            "Taking things slowly might help manage this fear. Try breaking down any challenges you face into smaller steps."
+        ],
+        "sadness": [
+            "Sadness can feel heavy. Maybe taking a walk or engaging in a comforting activity could lighten things a bit.",
+            "Remember, you don’t have to go through this alone. Reaching out to someone close might provide some comfort."
+        ],
+        "surprise": [
+            "Surprises can bring unexpected emotions. Take some time to process it and think about how you might want to respond.",
+            "New things can be a little unsettling, but they might also lead to positive changes. How do you feel about this surprise?"
+        ],
+        "neutral": [
+            "I’m here whenever you’d like to share more.",
+            "Sometimes, neutral moments give us a chance to recharge. Feel free to tell me anything on your mind."
+        ]
+    }
+
+
+    return random.choice(reframing_messages.get(emotion, ["I'm here to support you. Feel free to share more."]))
+
+def handle_emotion_analysis(input_text):
+    """Handle emotion and generate a response based on the detected emotion."""
+    emotion = analyze_emotion(input_text)
+    print("Detected Emotion:", emotion)  
+
+
+    reframing_message = generate_reframing_message(emotion)
+    response = f"{reframing_message} Is there anything else you’d like to discuss?"
+
+    return response
+
+def chatbot(text):
+    """Chatbot function to process the input and generate appropriate responses."""
+    if text:
+
+        emotion = analyze_emotion(text)
+        print("Detected Emotion:", emotion) 
+
+
+        reframing_response = generate_reframing_message(emotion)
+        reply = f"{reframing_response} Is there anything else you’d like to discuss?"
+
+        return reply
+    return "I'm here to help with your thoughts!"

ML/static/js/script.js

@@ -0,0 +1,21 @@
+document.getElementById("send-btn").addEventListener("click", function() {
+  const userInput = document.getElementById("user-input").value;
+
+  // Send a POST request to the /chat endpoint
+  fetch("/chat", {
+    method: "POST",  // Ensure POST is used, not GET
+    headers: {
+      "Content-Type": "application/json"
+    },
+    body: JSON.stringify({ messageText: userInput })  // Send the user input as JSON
+  })
+  .then(response => response.json())
+  .then(data => {
+    const chatBox = document.getElementById("chat-box");
+    const userMessage = `<div class="user-msg">${userInput}</div>`;
+    const botReply = `<div class="bot-msg">${data.answer}</div>`;
+    chatBox.innerHTML += userMessage + botReply;
+    document.getElementById("user-input").value = "";  // Clear input field after sending
+  })
+  .catch(error => console.error("Error:", error));
+});