🛡️ TriGuard: AI-Powered Anomaly Detection for Smart Cities

This project is a submission for the hackathon challenge to ensure data integrity in smart city environmental monitoring systems.

---

## 1. Problem Statement

The data from environmental sensors is the lifeblood of a smart city, but it can be manipulated. The task is to build an unsupervised machine learning model that continuously analyzes data streams from a network of sensors, learns normal operational patterns, and identifies subtle anomalies that could indicate a malfunctioning sensor or a sophisticated data tampering attack.

---

## 2. Our Solution - TriGuard

TriGuard is a full-stack application featuring a hybrid AI engine designed to detect a wide range of anomalies in real-time.

• Hybrid AI Engine: Utilizes three distinct unsupervised models (Isolation Forest, Autoencoder, and LSTM Autoencoder) to detect statistical, contextual, and temporal anomalies.
• Real-time API: A Flask backend serves the trained models and provides a live API endpoint for predictions.
• Data Integrity: Includes a (simulated) blockchain logging feature to create a tamper-proof audit trail of detected anomalies.

---

## 3. Tech Stack

• Backend: Python, Flask, Flask-CORS
• Machine Learning: TensorFlow (Keras), Scikit-learn, Pandas
• Frontend: React.js
• Deployment Goal: Backend on Render, Frontend on Vercel

---

## 4. Project Structure

TriGuard_Final/
│
├── backend/
│   ├── models/
│   ├── app.py
│   └── requirements.txt
│
└── frontend/
    ├── src/
    ├── public/
    └── package.json

---

## 5. Setup and Running Instructions

### Backend Server

1. Navigate to the backend directory:

   cd backend

2. Install the required Python packages:

   pip install -r requirements.txt

3. Run the Flask server:

   python app.py

   The server will be running on http://127.0.0.1:5000.

### Frontend Application

1. Navigate to the frontend directory in a new terminal:

   cd frontend

2. Install the required Node packages:

   npm install

3. Run the React development server:

   npm run dev

   The application will be accessible in your browser, likely at http://localhost:5173.