Publication

Abstract : Numerous elements, like the duration, place of travel, the time of purchase, etc., have an impact on the cost of an airline ticket. Each carrier has its own set of exclusive guidelines and the algorithms to demonstrate the appropriate price. It is now possible to deduce such rules and predict the price variance through recent advancements in artificial intelligence (AI) and machine learning (ML) methods. This paper proposes machine learning regression method to predict the flight rate from Bangalore to Kolkata utilizing real life data. The proposed framework uses various factors affecting flight rates (Date, Departure Time, Duration and No. of stops) as input and predict the flight price using six different regression methods, LGBM, Gradient Booster, XGB, Linear, SVR and MLP. By the analysis of the outcome's errors using several statistical measures, top efficient regression methods have been selected. The model has a high level of prediction accuracy with 0.065 RMSE and 0.04 MAE, which ensured the good efficiency of the model.