Predicting Flight Tickets using Gradient Boosting Regressor vs. Linear Regression

Abstract

This project's overarching goal is to use Gradient Boosting to make airfare predictions for future ticket purchases. In contrast to fresh linear regression, regression device learning is a collection of guidelines. Here are the materials and methods used: the New Linear Regression Algorithm and Gradient Boosted Regression. The sample size was 10. Two companies use this image to compute these algorithms, which are based on a total of twenty algorithm samples. A G Power value of 80% was used to compare the sample to a control group, and a sample size of 10 was calculated. Gradient Boosting regression (82.5% accuracy) trumps new Linear Regression (62.5%) in terms of attained values. 8%. Reason being, compared to new linear regression, Gradient Boosting regression yields better results. A one-tailed test revealed that the new linear regression technique has a 0.00% statistically significant difference from the Gradient Boosting Regressor. The significance level at which this result was reached was p 0.05. Finally, after analyzing all of the approaches, it was found that the airfare prediction outperformed the new linear regression. After completing all of the steps, this became apparent.