A machine learning project designed to predict flight ticket prices based on historical data and flight details. This project builds and deploys a predictive model, offering users an intuitive web interface for obtaining price estimates.
- Data Exploration: Analyze and visualize flight price datasets to uncover patterns and insights.
- Model Building: Train machine learning models using features like airline, source, destination, journey dates, and departure times to predict flight prices. Evaluate and compare model performance.
- Web Application: Deploy the trained model via a simple Flask web app, enabling users to interact and obtain predictions effortlessly.
data_exploration.ipynb: Notebook for dataset analysis and visualization.model_building.ipynb: Notebook for training and evaluating machine learning models.app.py: Flask application that serves predictions from the trained model.
- Interactive Web Interface: Enter flight details (e.g., airline, source, destination, journey date) to get price predictions.
- Customizable Inputs: Tune predictions with optional additional flight details.
- Accurate Modeling: Leverages advanced machine learning techniques for reliable price estimates.
This project uses the following libraries:
- Python 3
- NumPy
- Pandas
- Scikit-Learn
- XGBoost
- Flask
Refer to requirements.txt for a complete list of dependencies.
- Clone the repository and navigate to the project directory.
- Install dependencies:
pip install -r requirements.txt
- Start the Flask web app:
python app.py
- Open the application in your browser at
http://localhost:5000.
Provide the following mandatory details for accurate predictions:
- Airline
- Source
- Destination
- Journey Date
- Departure Time
Additional flight details can further enhance prediction accuracy.
Contributions are welcome! You can help improve:
- Model accuracy and feature selection.
- User interface of the web app.
- Bug fixes and documentation enhancements.
Developed by Taaran Jain.
Feel free to submit issues or pull requests to collaborate on this project.