nguyetvo / cbd_airlines-arrival Goto Github PK

Airline Arrivals Project

Use this dataset of airline arrival information to predict how late flights will be. A flight only counts as late if it is more than 30 minutes late.

1. The project should follow guideline as previous projects.
2. Apply models in Naïve Bayes, Logistic Regression, Decision Tree, Random Forest, Gradient Boosting and SVM.
3. Apply PCA, SelectKBest and RFE for feature selections.
4. Using gridsearchCV to obtain best parameters for models.
5. Compare performances among models, write up analysis why the model is good or bad in the algorithmic approach (explain why the algorithm is good or bad for the dataset structure, can you do something to improve the model?)
6. Include the conclusions.