Intelligent Processing of Intrusion Detection Data

Tao Duan,Youhui Tian,Jian Jiang,Qianmu Li,Yaozong Liu,Hanrui Zhang,Zongsheng Shi

doi:10.1109/access.2020.2989498

Abstract

Intrusion detection technology, as an active and effective dynamic network defense technology, has rapidly become a hot research topic in the field of network security since it was proposed. However, current intrusion detection still faces some problems and challenges that affect its detection performance. Especially with the rapid development of the current network, the volume and dimension of network data are increasing day by day, and the network is full of a large number of unlabeled data, which brings great pressure on the data processing methods of IDS. In view of the tremendous pressure of intrusion detection brought by the current complex and high-dimensional network environment, this paper provides a feasible solution. Firstly, this paper briefly outlines the necessity of feature learning, the shortcomings of traditional feature learning methods and the new breakthroughs brought by deep belief network in feature learning, and focuses on the principle and working mechanism of deep belief network and Principal Component Analysis (PCA). Then, it constructs the intrusion detection model based on PCA-BP and DBN respectively. And through the experimental evaluation of the two detection models, a comparative experiment between deep belief network and principal component analysis is constructed. The experimental results show that deep belief network has unique advantages and good performance in feature learning. Therefore, deep belief network can be applied in the field of intrusion detection to extract effective features from the current high-dimensional and redundant network data, thereby improving the detection performance of IDS and its adaptability to the current complex and high-dimensional network environment.

Highlights

In the current stage of intrusion detection research, integrating intelligent technologies such as expert systems, statistical analysis, and data mining into intrusion detection has become a hot topic in the field of intrusion detection [1]–[5]
ALGORITHMIC IDEAS The basic idea of intrusion detection data processing based on Principal Component Analysis (PCA)-Back Propagation (BP) neural network is: First, the high-dimensional standard to-be-detected data generated by the preprocessing module is subjected to feature learning through the PCA feature learning module to eliminate redundant information
WORKING STEPS The working steps of the intrusion detection data processing model based on PCA-BP neural network are as follows: Model training phase: Step 1: Pre-process the labeled training data to obtain high-dimensional labeled training with standardized format data; Step 2: Use PCA to perform feature learning on the preprocessed training data to eliminate redundant and useless information

Summary

INTRODUCTION

In the current stage of intrusion detection research, integrating intelligent technologies such as expert systems, statistical analysis, and data mining into intrusion detection has become a hot topic in the field of intrusion detection [1]–[5]. Before the intrusion detection model performs feature learning, it becomes a necessary process to obtain better low-dimensional feature information of the original data [22]. B. ALGORITHMIC IDEAS The basic idea of intrusion detection data processing based on PCA-BP neural network is: First, the high-dimensional standard to-be-detected data generated by the preprocessing module is subjected to feature learning through the PCA feature learning module to eliminate redundant information. C. WORKING STEPS The working steps of the intrusion detection data processing model based on PCA-BP neural network are as follows: Model training phase: Step 1: Pre-process the labeled training data to obtain high-dimensional labeled training with standardized format data; Step 2: Use PCA to perform feature learning on the preprocessed training data to eliminate redundant and useless information. The judgment principle here is the same as the BP network classifier in intrusion detection data processing based on PCA-BP

FEATURE LEARNING COMPARISON EXPERIMENT

Findings

CONCLUSION