Research on Network Intrusion Detection Based on an Improved Deep Learning Method

Abstract

Network intrusion detection is an important research direction in the field of network security. The traditional detection algorithm is based on feature extraction and feature separation, which has the problems of low detection accuracy and high false alarm rate. In order to improve the accuracy of network intrusion detection, this paper proposes an intrusion detection model based on deep asymmetric convolutional encoder and Random Forest(RF). First, use DACAE to extract features from the preprocessed data, and then use the random forest algorithm to divide the network traffic data into normal and abnormal classes, and finally achieve the purpose of network intrusion detection. It is tested on three public benchmark datasets of network intrusion detection NSL-KDD and KDD99 datasets. The experimental results show that the accuracy and false alarm rate of the improved method are better than the comparative method.