Network intrusion detection method and system based on federated learning and block chain

The invention belongs to the field of network security, and provides a federated learning and block chain-based network intrusion detection method and system, and the method comprises the steps: obtaining network intrusion data, and carrying out the data preprocessing; performing feature extraction based on the preprocessed network intrusion data to obtain network intrusion features; according to the network intrusion features, performing network intrusion detection by using a pre-trained network intrusion detection classifier; wherein the network intrusion detection classifier is a deep forest classifier based on a fruit fly optimization algorithm, and during training, the fruit fly optimization algorithm is adopted to optimize two parameters of the number of forest trees and the number of leaf nodes of the deep forest classifier; and obtaining a deep forest classifier based on the fruit fly optimization algorithm based on the number of the forest trees and the number of the leaf nodes after optimization. Ac