Key Feature Recognition Algorithm of Network Intrusion Signal Based on Neural Network and Support Vector Machine

Abstract

When identifying the key features of the network intrusion signal based on the GA-RBF algorithm (using the genetic algorithm to optimize the radial basis) to identify the key features of the network intrusion signal, the pre-processing process of the network intrusion signal data is neglected, resulting in an increase in network signal data noise, reducing the accuracy of key feature recognition. Therefore, a key feature recognition algorithm for network intrusion signals based on neural network and support vector machine is proposed. The principal component neural network (PCNN) is used to extract the characteristics of the network intrusion signal and the support vector machine multi-classifier is constructed. The feature extraction result is input into the support vector machine classifier. Combined with PCNN and SVM (Support Vector Machine) algorithms, the key features of network intrusion signals are identified. The experimental results show that the algorithm has the advantages of high precision, low false positive rate and the recognition time of key features of R2L (it is a common way of network intrusion attack) data set is only 3.18 ms.Symmetry 2019, 11, 380 2 of 15 features and remove secondary features, which is often conducive to shortening the detection time and discovering the intrinsic features of a certain type of attack [10]. Feature extraction is not to explicitly remove some features from the input features, but to carry out linear or nonlinear transformation of the input features, extracting from them to replace the table components, using these components instead of the original input features, so as to achieve dimensionality reduction and feature space conversion [11][12][13].At present, there are few researches on intrusion detection from the perspective of feature extraction. Authors in [14] constructed a multi-layer hybrid intrusion detection model, using support vector machine and extreme learning machine to improve the efficiency of detection of known and unknown attacks; then proposed an improved k-means method, established a new small training data set representing the entire training data set, greatly shortened the training time of classifier, and improved the performance of the intrusion detection system. However this method has the problem of low accuracy. Authors in [15] constructed a wireless grid intrusion detection system based on genetic algorithm and multi-support vector machine classifier. The system chooses the information characteristics of each type of attack, rather than the common features of all attacks. The network simulator is used to simulate the intrusion data set generated by wireless mesh network. The system is evaluated with the parameters of packet transmission rate and delay. However, the system has the problem of high false alarm rate.It can be mentioned that dimensionality reduction in NN inputs and topology to enhance generalisation has been applied in other fields. Authors in [16] constructed a spectrum prediction method based on back propagation-training...

A Cloud Load Forecasting Model with Nonlinear Changes using Whale Optimization Algorithm Hybrid Strategy - extreme Learning Machine

“…Cloud computing is applied to process massive data in a short time on the distributed parallel computing basis, and has strong network service ability (Kim and Jeong, 2017). Achieving efficient resource management has always been the most concerned issue for cloud service operators and providers (Khalilpourazari and Khalilpourazary, 2018;Ye et al, 2019;Mehrabi et al, 2021). Cloud load prediction is necessary for efficient network regulation in the cloud environment (Xu et al, 2017).…”

SVM-Lattice: A Recognition and Evaluation Frame for Double-Peaked Profiles

“…The SVM algorithm also is used as a powerful classification tool for cancer genomic classification [31]. In addition to the single algorithm, a combination classification schema of SVM with other methods is applied to improve convective and stratiform classification, including random forest [32], logistic repression [33], neural network [34], etc. For the popularization of high-dimensional data, the SVM is employed to construct a classification model based on the feature selection or extraction method [1], [35], [36].…”

A Multichannel Data Fusion Method Based on Multiple Deep Belief Networks for Intelligent Fault Diagnosis of Main Reducer

“…Time domain analysis technique is easy to implement and independent to rotating speed. Frequency domain analysis technique is superior to time domain analysis technique for early-stage and distributed faults [28][29][30][31]. Therefore, we employ eight statistical parameters including kurtosis, crest factor, variance, standard deviation, RMS, skewness, mean and impulse indicator [32] expressed as follows to implement the preprocessing of time domain and frequency domain signals.…”