Flight Delay Classification Prediction Based on Stacking Algorithm

With the development of civil aviation, number flights keeps increasing and flight delay has become a serious issue even tends to normality. This paper aims prove that Stacking algorithm advantages in airport prediction, especially for selection problem machine learning technology. In this research, principle classification is introduced, SMOTE selected process imbalanced datasets, Boruta utilized feature selection. There are five supervised algorithms first-level learner including KNN, Random Forest, Logistic Regression, Decision Tree, Gaussian Naive Bayes. The second-level Regression. To verify effectiveness proposed method, comparative experiments carried out based on Boston Logan International Airport datasets from January December 2019. Multiple indexes used comprehensively evaluate prediction results, such as Accuracy, Precision, Recall, F1 Score, ROC curve, AUC Score. results show not only could improve accuracy but also maintains great stability.