Network Intrusion Detection Method Based on Stack Denoising Auto-Encoder and Gradient Boosting Decision Tree

• 1、College of Computer Science and Electronic Engineering, Hunan University, Changsha, 410082

Abstract：With the rapid development of network application technology, hidden dangers of network security are increasing day by day, and large-scale network attack events often occur, which put forward higher requirements for the development of network security technology. Aiming at the problems of low detection performance and high false detection rate of existing network intrusion detection methods when processing high-dimensional massive non-linear data, a stack denoising auto-encoder (SDAE) is proposed Combined with Gradient Boosting Decision Tree (GBDT), the combined intrusion detection model (SDAE-GBDT) uses the SDAE encoder to extract the deep features of network traffic data, and combines the good classification performance of GBDT to achieve high accuracy intrusion detection.

Keywords： computer technology network intrusion detection SDAE encoder deep learning gradient boosting decision tree integrated learning