Flight Delay Prediction Using Different Regression Algorithms in Machine Learning

The two forms of regression algorithms are investigated and compared in this Article Such as LASSO and RIDGE regression. For scheduled airlines to improve customer satisfaction and success, accurate forecasting of delay is essential. There is no way to stop a flight from being delayed, yet they significantly affect the profits and losses of carriers. This research investigates a larger spectrum of potential flight delay difficulties and compares two machine learning algorithms in defined extended flight delay time series forecasting. A dataset for the suggested technique is created by gathering, decoding, and linking automatic dependent surveillance-broadcast (ADS-B) signals with additional data including weather, flight schedules, and airport information. A regression approach is used in conjunction with a number of forecasting tasks as part of the defined prediction challenges. The accuracy of the suggested prediction model was examined and compared to current prediction approaches. The results of LASSO and RIDGE regression with the mean absolute errorMAE (0.2 and 0.1), mean squared error-MSE (0.1 and 0.04), root mean square error-RMSE (0.3 and 0.2) and Accuracy (99.7% and 99.8%) respectively.