This repository contains all the files for final projects.
Analysis of flight status is important for both flight companies and clients. For companies, sufficent advance preparations before the flight delay high-occurrence seasons are helpful in flight organization and emergence prevention. For clients and customers, an advance estimation about the flight delay or cancellation could give them advice on schdual planning and insurrance purchase. For this project, I plan to use the U.S. flgiht database of year 2008 to predict whether a flight would be on-time or delayed or even cancelled according to the flgiht information such as dates, carriers and schedueled departure time.
1.Basic information about the flight dataset.
2.Define the prediction target---the flight status.
3.Select relative features.
4.Make prediction using supervised learning models.
5.Analyze how selection of features may affect the selection of models.
The database is about the on-time performance of the domestic flgiht provided by The U.S. Department of Transportation's (DOT) Bureau of Transportation Statistics (BTS). The database is open source on Kaggle.
Detail information can be found here:
https://www.kaggle.com/giovamata/airlinedelaycauses
The code in this repository is released under the MIT license. Read more at the Open Source Initiative.