Prediction of Optimal Flight Ticket Purchase Timing by Using Learning To Rank Methods

Airline ticket prices vary according to a lot of factors such as the flight day, departure and arrival cities, flight duration, or days remaining to the flight. Airline companies make pricing based on the algorithms they have developed. On the customer side, there are two prominent issues to be addressed: price estimation and optimal purchase date prediction. Although the former has been studied a lot, there are a few studies for the latter. By using Learning to Rank methods, recommendation models can list the optimal dates for purchasing the flight ticket. This study aims to analyze the effect of the number of days remaining on the price based on the flight route by creating a dataset of domestic flights and developing a recommendation algorithm for optimal purchase dates.