A Network Intrusion Detection Method for Training Dataset Generation Based on KNN

In this paper, for overcoming the difficulty of training dataset about two random method and manual method, the K-Nearest Neighbor (KNN) method is used to detect the network intrusion. Through adjusting K-value from part dataset and overall dataset, it can be calculated that the accuracy rate of random method is higher than of manual method for network intrusion detection by KNN. Finally, based on CICIDS2017 dataset, the experiments can verity the effectiveness of the proposed method.