FLIGHT DELAY PREDICTION MODEL BASED ON FEDERATED LEARNING

In view of the fact that the existing flight delay prediction methods do not consider the problems of multiple data sources and data privacy,this paper proposes a federated learning framework,which integrates logistic regression methods,so that the training data can kept locally without uploading and sharing,and the flight delay can be predicted on the premise of protecting data privacy.At the same time,aimed at the problem of indirect information leakage in the training process,homomorphic encryption technology was adopted to encrypt the transmitted parameters.The experimental results show that the federated modeling method can achieve similar accuracy than the traditional method without sharing data,which provides a practical scheme for optimizing civil aviation business.