Open AccessA Prediction Method of Network Security Situation based on QPSO-SVM
Author(s) -
Jianan Zhang,
Hui Luo
Publication year - 2020
Publication title -
international journal of circuits, systems and signal processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.156
H-Index - 13
ISSN - 1998-4464
DOI - 10.46300/9106.2020.14.105
Subject(s) - support vector machine , computer science , particle swarm optimization , adaptability , sequence (biology) , intrusion detection system , data mining , scalability , artificial intelligence , machine learning , network security , face (sociological concept) , artificial neural network , computer security , ecology , social science , genetics , database , sociology , biology
In network security situation awareness system, situation prediction is the key point. The traditional intrusion detection method lacks scalability in the face of the changing network structure and lacks adaptability in the face of unknown attack types. In order to ensure and improve the accuracy of situation prediction, a QPSO-SVM prediction model is proposed by combining the optimization performance of quantum particle swarm optimization and the prediction accuracy of support vector machines. By adding the original sequence to the original sequence, this model weakens the irregular disturbance in the original sequence and enhances the regularity of the sequence. Compared with the traditional SVM and PSOSVM, the superiority of the prediction precision is better, the prediction accuracy can be ensured, and the validity of the model is tested by the simulation experiment.