FLIGHT TICKET PRICE PREDICTOR

Abstract

People who frequently travel by flight usually have better knowledge about discounts and the right time to purchase tickets. For business purposes, many airline companies adjust ticket prices depending on seasons and time periods. Prices generally increase during times when more people travel. To estimate airline ticket prices for a particular route, data is collected with features such as Duration, Source, Destination, Arrival, and Departure. These features are taken from a selected dataset, and in this study machine learning techniques and regression methods are used to predict ticket prices, as airline fares vary over time. We implemented a flight price prediction system for users using Decision Tree and Random Forest algorithms. Among these, the Decision Tree algorithm achieved the highest accuracy of about 80% in predicting flight prices. In addition, correlation analysis and ANOVA tests were performed to support the statistical evaluation of the model.