Title: Airline dynamic price prediction using machine learning

Authors: Kushal Kumar Ruia; Utkarsh Daga; Aditya Tripathi; Maruf Nissar Rahman; Saurabh Bilgaiyan

Addresses: School of Computer Engineering, KIIT, Deemed to be University, Bhubaneswar, India ' School of Computer Engineering, KIIT, Deemed to be University, Bhubaneswar, India ' School of Computer Engineering, KIIT, Deemed to be University, Bhubaneswar, India ' School of Computer Engineering, KIIT, Deemed to be University, Bhubaneswar, India ' School of Computer Engineering, KIIT, Deemed to be University, Bhubaneswar, India

Abstract: Nowadays, to optimise revenue, airlines use dynamic pricing techniques. Earlier demand was predicted by retrospective analysis of sales data across various sales channels. This technique becomes less reliable when the volatility of the market and competition increases. In this paper, different models have been built using machine learning techniques like regression which could predict the ticket prices of airlines based on features such as journey route, historical ticket price, etc. Along with this, the dependency of important features on accuracy has also been studied. An experimental study of this paper reveals that regression algorithms can be used to predict airline price as the highest R² score achieved on testing data is 0.85 using gradient boosting regressor. The objective of this paper is to understand the key features that affect the airline price and compare the accuracy of the different machine learning algorithms with the selection of different features.

Keywords: machine learning; prediction model; air tickets prices; regression; R² score; cross-validation.

DOI: 10.1504/IJPQM.2022.124712

International Journal of Productivity and Quality Management, 2022 Vol.36 No.2, pp.187 - 207

Received: 29 Jul 2020
Accepted: 05 Jan 2021
Published online: 08 Aug 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article