AI-Powered Superannuation Advisor Dashboard

A comprehensive AI-powered superannuation advisor with ML models, real-time predictions, conversational AI, Azure Speech Services, and automated email system. Features a clean, card-based, accessible dashboard UI targeted at older users with high-contrast design, large fonts, and simple navigation.

🚀 Key Features

📊 Dashboard Overview

• User Selection: Dropdown with 10 sample User_IDs from the dataset for testing different user profiles
• Summary Card: Shows retirement amount, monthly increase needed, and peer comparison
• Key Metrics: Current balance, % to goal, estimated monthly income at 65, contributions
• Peer Comparison: Investment strategy comparison with age/risk groups

📈 Portfolio Management

• Asset Allocation Cards: Visual breakdown by asset type (Stocks, ETFs, Managed Funds, etc.)
• Donut Chart: Interactive asset allocation with hover tooltips
• Growth Projection: Line chart showing investment growth with milestones
• Investment Types: Display of user's current investment types and funds

🎯 Goals Planning

• Milestone Cards: Display user goals with progress tracking
• Dynamic Goal Addition: Add short-term and long-term goals
• Progress Tracking: Visual progress bars and completion status
• Retirement Impact: Shows how goals affect retirement projections

📚 Education & Benefits

• Tax Benefits: Explanation of superannuation tax advantages
• Government Pension: Age Pension eligibility and benefits
• Private Pension: Private pension options and strategies
• Withdrawal Strategies: Fixed, Dynamic, and Bucket strategies
• "Explain This" Buttons: Chatbot integration for plain language explanations

⚖️ Risk Management

• Risk Tolerance Slider: Low/Medium/High risk selection
• Asset Allocation Input: Interactive sliders for ideal allocation
• Comparison Tool: Compare user allocation with recommended allocation
• Risk Assessment: Comprehensive risk analysis and recommendations

🤖 AI Chatbot with Voice

• Floating Action Button: Easy access to AI advisor
• Voice Interaction: Text-to-speech capabilities
• Voice Controls: Start/stop recording, voice selection, enable/disable
• Individual Message Playback: Speaker icons for manual message reading
• Sample Questions: Pre-built questions for common scenarios
• ML Model Integration: Simulated responses based on user data
• Contextual Advice: Personalized recommendations based on user profile

🎤 Azure Speech Services

• Speech-to-Text (STT): Voice input for chatbot interactions
• Multiple Voice Options: Australian, US, and UK English voices
• Voice-Guided Signup: Voice-assisted user profile creation
• Real-time Audio Processing: Optimized for 16kHz, 16-bit, mono audio

📧 Automated Email System

• 6-Hour Scheduled Emails: Automated financial updates every 6 hours
• Manual Email Trigger: "Get Financial Update" button with news source selection
• News Source Options:
  • 📰 NewsAPI: Real-time financial news from NewsAPI.org
  • 🤖 Gemini AI: AI-generated financial content with recent web data
• Professional HTML Templates: Beautiful, responsive design with MUFG branding
• Comprehensive Content Categories:
  • 📈 Stock Market & Economic Updates
  • 🏠 Real Estate & Bonds
  • 📋 Policy Changes & Tax Updates
  • 📰 General Financial News

📁 Project Structure

├── backend/               # Python ML backend
│   ├── api.py            # FastAPI server
│   ├── train.py          # ML model training
│   ├── chat_router.py    # AI chatbot integration
│   ├── inference.py      # ML model inference
│   ├── requirements.txt  # Python dependencies
│   └── setup.bat/sh      # Backend setup scripts
├── src/
│   ├── components/
│   │   ├── dashboard/    # Dashboard components
│   │   └── ui/           # Reusable UI components
│   ├── pages/            # Main application pages
│   ├── services/         # Data service layer
│   └── lib/              # Utility functions
└── case1.csv             # Dataset

🛠️ Technical Implementation

Data Service

• Mock Data: 10 sample users with realistic financial profiles
• Peer Comparison: Age group and risk group comparisons
• Projection Calculations: Retirement amount and contribution calculations
• Investment Types: Support for various Australian investment categories

Components Structure

src/
├── components/dashboard/
│   ├── UserSelection.tsx          # User profile switcher
│   ├── DashboardHeader.tsx        # User info and progress
│   ├── SummaryCard.tsx            # Retirement projection
│   ├── MetricsGrid.tsx            # Key financial metrics
│   ├── NavigationTabs.tsx          # Main navigation
│   ├── PortfolioPage.tsx          # Asset allocation & growth
│   ├── GoalsPage.tsx              # Financial goals management
│   ├── EducationPage.tsx          # Benefits & education
│   ├── RiskPage.tsx               # Risk assessment
│   ├── ChatbotPage.tsx            # AI advisor interface
│   └── FloatingChatButton.tsx     # Quick chat access
├── services/
│   └── dataService.ts            # Data management & calculations
└── pages/
    └── Dashboard.tsx              # Main dashboard component

Key Technologies

• React 18: Modern React with hooks
• TypeScript: Type-safe development
• Tailwind CSS: Utility-first styling
• Radix UI: Accessible component primitives
• Recharts: Data visualization
• Lucide React: Consistent iconography

📱 User Experience

For Older Users

• Large, Clear Text: Easy-to-read fonts and high contrast
• Simple Navigation: Intuitive tab-based interface
• Visual Feedback: Clear status indicators and progress bars
• Helpful Tooltips: Contextual information and explanations
• Consistent Layout: Predictable interface patterns

Financial Planning Features

• Scenario Planning: "What if" calculations for contributions
• Risk Assessment: Personalized risk tolerance evaluation
• Goal Tracking: Visual progress toward financial goals
• Peer Comparison: Benchmark against similar users
• Educational Resources: Built-in learning materials

🚀 Getting Started

Quick Setup (Automated)

Windows:

cd backend
setup.bat

Linux/Mac:

cd backend
chmod +x setup.sh
./setup.sh

Manual Setup

1. Install Python Dependencies

   cd backend
   pip install -r requirements.txt

2. Install Node.js Dependencies in the ROOT Folder

   npm install

3. Train ML Models

   cd backend
   python train.py

4. Set up Environment Variables Create a .env file in the backend folder with your API keys:

   # Hugging Face Token (for ML models)
   HF_TOKEN=your_huggingface_token_here
   
   # Azure Speech Services (for voice features)
   AZURE_SPEECH_KEY=your_azure_speech_key_here
   AZURE_SPEECH_REGION=your_azure_region_here
   
   #Supabase
   SUPABASE_URL=
   SUPABASE_SERVICE_KEY=
   VITE_SUPABASE_URL=
   VITE_SUPABASE_ANON_KEY=
   
   # Email Service (for automated emails)
   SMTP_HOST=smtp.gmail.com
   SMTP_PORT=587
   SMTP_SECURE=false
   SMTP_USER=your_email@gmail.com
   SMTP_PASSWORD=your_app_password
   ADMIN_EMAIL=your_email@gmail.com
   
   # News APIs (optional - for email content)
   NEWS_API_KEY=your_news_api_key_here
   GEMINI_API_KEY=your_gemini_api_key_here

5. Start the ML Backend

   cd backend
   python api.py

6. Start the Frontend

   npm run dev

7. Access the Application

   • Frontend: http://localhost:8080
   • Backend API: http://localhost:8000
   • Health Check: http://localhost:8000/health

Testing Different Users

Use the user selection dropdown to switch between different user profiles (U1000-U1009)

🤖 ML Models & AI Features

Machine Learning Models

• KMeans Clustering: User segmentation based on demographics and financial behavior
• Logistic Regression: Risk tolerance prediction using financial and demographic features
• XGBoost: Investment recommendations and pension projections

AI Chatbot Integration

• Mistral LLM: Conversational AI powered by meta-llama/Llama-3-8b-chat
• Context-Aware Responses: AI pulls user data from ML models and CSV dataset
• Natural Language Processing: Understands queries about risk, contributions, projections, and peer comparisons

API Endpoints

Core ML & Data Endpoints

• /user/{user_id} - Get user profile
• /predict/{user_id} - Pension projection + risk category
• /summary/{user_id} - Dashboard summary statistics
• /peer_stats/{user_id} - Peer comparison data
• /simulate - Run projections with adjusted contributions
• /risk/{user_id} - Risk tolerance prediction
• /segment/{user_id} - User segmentation
• /recommendations/{user_id} - Investment recommendations

Azure Speech Services Endpoints

• POST /text-to-speech - Convert text to speech
• POST /speech-to-text - Convert speech to text
• GET /voices - Get available voices

Email System Endpoints

• GET /trigger-email - Trigger email with default source
• GET /trigger-email/{news_source} - Trigger email with specific source (gemini/newsapi)
• POST /send-email - Send email with custom parameters

System Endpoints

• GET /health - Check system status including email scheduler

📊 Dataset Integration

The application uses real data from case1.csv with 500+ users including:

• Demographics (Age, Gender, Country, Employment Status)
• Financial Data (Income, Savings, Contributions, Investment Types)
• Risk Profiles and Investment Experience
• Pension Eligibility and Tax Benefits
• Transaction History and Portfolio Diversity

🎯 How to Use New Features

Voice-Enabled Chatbot

1. Go to the "Chatbot" tab in the dashboard
2. Click "Start Recording" to speak your question
3. AI will respond with both text and voice
4. Use speaker icons next to messages for manual playback
5. Select different voices from the dropdown

Voice-Guided Signup

1. When creating a new user profile, use the enhanced signup form
2. Enable "Voice Guided Mode" for step-by-step voice assistance
3. Speak your answers to questions
4. The form will be filled automatically based on your speech input

Automated Email System

1. Emails are sent automatically every 6 hours
2. Use the "Get Financial Update" button in the dashboard header
3. Select news source (NewsAPI or Gemini AI) before sending
4. Check your email for comprehensive financial updates

🔮 Future Enhancements

• Real Excel Integration: Connect to actual dataset
• Real-time Data: Live market data integration
• Advanced Analytics: More sophisticated financial modeling
• Mobile App: Native mobile application
• Enhanced Voice Commands: More natural voice interactions

🖼️ Preview

• Landing Page

• Signup Page

• User Dashboard

🚀 Live Demo

📄 License

This project is part of the MUFG Hackathon and is designed for demonstration purposes.

---