Predicting Airline Ticket Prices Using Decision Trees and Random Forests: A Comparative Analysis

In the current era of abundant data, predicting fluctuating prices of resources such as airline tickets remains a challenging task without the appropriate analytical tools. This study explores the use of decision tree-based models, including random forests, to predict airline ticket prices. I hypothesize that random forests, due to their ability to reduce over-fitting, will outperform single decision trees in this task. my approach is validated using a dataset of over 300,000 data points from "Ease My Trip," collected over 50 days. The results demonstrate that random forests achieve a lower root mean squared error (RMSE) compared to single decision trees, confirming the robustness of ensemble methods in improving prediction accuracy. Additionally, my analysis highlights the importance of model selection and hyper-parameter tuning in achieving optimal performance. This paper aims to provide a robust and empirically tested procedure for flight ticket price prediction, with potential applications in dynamic pricing strategies for airlines and cost-saving tools for travelers. Future work will explore other ensemble methods, such as gradient boosting and probabilistic regression trees, to further enhance prediction accuracy