🛸 UFO Sighting Predictor

A machine learning-powered web application that predicts the country where a UFO sighting was most likely reported based on duration, latitude, and longitude data.

📋 Overview

This web application uses a trained logistic regression model to predict which country a UFO sighting was reported in, based on three key features:

• Duration (in seconds, 0-60)
• Latitude (geographical coordinate)
• Longitude (geographical coordinate)

The model can predict sightings from 5 countries: Australia, Canada, Germany, United Kingdom, and United States.

✨ Features

• 🌐 Interactive Web Interface: Clean, modern UI with gradient text effects and smooth animations
• 🤖 Machine Learning Prediction: Trained logistic regression model using scikit-learn
• 📊 Real-time Results: Instant predictions based on user input
• 📱 Responsive Design: Works seamlessly across desktop and mobile devices
• ℹ️ Educational Content: Includes information about the model and how to use it

🛠️ Technologies Used

Backend

• Python 3.x: Core programming language
• Flask: Lightweight web framework for routing and serving the application
• scikit-learn: Machine learning library for model training and prediction
• NumPy: Numerical computing library for array operations
• Pandas: Data manipulation and analysis

Frontend

• HTML5: Semantic markup structure
• CSS3: Modern styling with gradients, animations, and glassmorphism effects
• JavaScript: Interactive button effects and dynamic user interactions
• Google Fonts: Inter and Playfair Display for premium typography

Machine Learning

• Model: Logistic Regression classifier
• Training Data: UFO Sightings dataset from Kaggle
• Features: Duration (seconds), Latitude, Longitude
• Target: Country of reported sighting (5 classes)

📁 Project Structure

ufo-webapp/
├── app/
│   ├── app.py                 # Main Flask application
│   ├── requirements.txt       # Python dependencies
│   ├── static/
│   │   ├── css/              # Stylesheets (global, layout, components)
│   │   └── js/               # JavaScript components (buttons, background)
│   └── templates/            # HTML templates (base, landing, index, about)
├── model/
│   └── ufo-model.pkl         # Trained ML model (pickled)
├── data/
│   └── raw/                  # Raw dataset storage
├── train.ipynb               # Jupyter notebook for model training
└── README.md                 # This file

🚀 Installation

Prerequisites

• Python 3.7 or higher
• pip (Python package manager)

Step-by-Step Setup

1. Clone the repository

   git clone <repository-url>
   cd ufo-webapp

2. Create a virtual environment (recommended)

   python3 -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies

   cd app
   pip install -r requirements.txt

▶️ Running the Application

1. Navigate to the app directory

   cd app

2. Start the Flask development server

   python app.py

3. Open your browser and navigate to:

   http://127.0.0.1:5000
   

4. Use the application

   • Visit the landing page at /
   • Navigate to the prediction form at /prediction
   • Enter duration (seconds), latitude, and longitude
   • Click "Predict!" to see which country the sighting likely occurred in

📖 Usage Example

To predict a UFO sighting in New York City:

1. Navigate to /prediction
2. Enter the following values:
   • Seconds: 30
   • Latitude: 40.7128
   • Longitude: -74.0060
3. Click Predict!
4. Result: United States 🇺🇸

🧠 Model Training

The model was trained using the train.ipynb Jupyter notebook. To retrain the model:

1. Open train.ipynb in Jupyter Notebook or JupyterLab
2. Ensure the UFO sightings dataset is in the data/raw/ directory
3. Run all cells to train the logistic regression model
4. The trained model will be saved as model/ufo-model.pkl

🙏 Acknowledgments

Thank you to my parents for always believing in me :D