Machine Learning-Based Flight Price Prediction System Using Linear Regression and GUI Integration

Abstract

The rapid growth of the aviation industry has led to dynamic and highly volatile flight pricing mechanisms influenced by multiple factors such as demand, seasonality, airline competition, and route characteristics. Predicting flight ticket prices accurately has become a crucial requirement for both customers seeking affordable travel and airlines aiming to optimize revenue. This project presents a machine learning-based Flight Price Prediction System that leverages historical flight data to estimate ticket prices efficiently. The proposed system uses Linear Regression, a supervised learning algorithm, to model the relationship between various input features and the flight price. The dataset is preprocessed using techniques such as label encoding for categorical variables and feature scaling for numerical consistency. Important features such as airline, source, destination, journey date, duration, and number of stops are extracted and transformed into meaningful numerical representations. These features are then used to train the regression model. To enhance usability, the system integrates a Graphical User Interface (GUI) built using Tkinter. This interface allows users to input flight details through dropdown menus and text fields, making the system interactive and user-friendly. Upon submission, the system processes the input data, applies the trained model, and displays the predicted price instantly. The system demonstrates how machine learning can be effectively applied to real-world problems involving dynamic pricing. While Linear Regression provides a simple and interpretable baseline model, the system architecture allows for future enhancements using advanced algorithms such as Random Forests or Gradient Boosting. Overall, this project highlights the importance of predictive analytics in the travel domain and provides a scalable solution that can be extended with real-time data and improved models. The integration of machine learning with a GUI ensures accessibility for nontechnical users, making the solution practical and impactful.