Flight prices fluctuate due to multiple factors including airline demand, booking time and route popularity. Predicting flight prices can help travelers make informed booking decisions and can help provide information about the most important factors that contribute to pricing.
This project aims to develop a neural networks model to predict flight ticket prices based on historical data. I will compare different neural network approaches as well as simpler baseline approaches like regressions to determine a model that is most suitable to making such a prediction. The project will involve a comparison as well as application approach.
I will evaluate these models on price prediction accuracy and explore key factors affecting airfare fluctuations, such as travel distance, departure time, and airline choices. The model can also be evaluated on real world data to test its accuracy and applicability.
Practical Application – Helps users predict the best time to book flights
Testing various models – Tests whether neural networks improve predictions
A successful project will: