Airline Delay Prediction using Machine Learning and Deep Learning Techniques
Devansh Shah1, Ayushi Lodaria2, Danish Jain3, Lynette D’Mello4
1Devansh Shah, Department of Computer Engineering, Dwarkadas J. Sanghvi College of Engineering, University of Mumbai, Mumbai, India.
2Ayushi Lodaria, Department of Computer Engineering, Dwarkadas J. Sanghvi College of Engineering, University of Mumbai, Mumbai, India.
3Danish Jain, Department of Computer Engineering, Dwarkadas J. Sanghvi College of Engineering, University of Mumbai, Mumbai, India.
4Lynette D’Mello, Asst. Prof., Department of Computer Engineering, Dwarkadas J. Sanghvi College of Engineering, University of Mumbai, Mumbai, India.
Manuscript received on May 25, 2020. | Revised Manuscript received on June 29, 2020. | Manuscript published on July 30, 2020. | PP: 1049-1054 | Volume-9 Issue-2, July 2020. | Retrieval Number: B4047079220/2020©BEIESP | DOI: 10.35940/ijrte.B4047.079220
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In this paper, we have tried to predict flight delays using different machine learning and deep learning techniques. By using such a model it can be easier to predict whether the flight will be delayed or not. Factors like ‘Weather Delay’, ‘NAS Delay’, ‘Destination’, ‘Origin’ play a vital role in this model. Using machine learning algorithms like Random Forest, Support Vector Machine (SVM) and K-Nearest Neighbors (KNN), the f1-score, precision, recall, support and accuracy have been predicted. To add to the model, Long Short-Term Memory (LSTM) RNN architecture has also been employed. In the paper, the dataset from Bureau of Transportation Statistics (BTS) of the ‘Pittsburgh’ is being used. The results computed from the above mentioned algorithms have been compared. Further, the results were visualized for various airlines to find maximum delay and AUC-ROC curve has been plotted for Random Forest Algorithm. The aim of our research work is to predict the delay so as to minimize loses and increase customer satisfaction.
Keywords: Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Random Forest, Long Short-Term Memory (LSTM), RNN.