A Novel Approach of KPCA and SVM for Intrusion Detection

Provided by: Nanjing University of Science & Technology
Topic: Security
Format: PDF
A novel hybrid approach of Kernel Principal Component Analysis (KPCA) and improved Support Vector Machine (SVM) using Genetic optimization Algorithm (GA) is proposed for intrusion detection. The original data is normalized preprocessing, KPCA is used as a preprocessor of SVM to extract the principal features of the normalized data, SVM is used to classification forecasting by finding the most appropriate kernel function and the optimal parameters with GA. Additionally, the proposed KPCA SVM with GA that can automatically determine the optimal parameters was tested on intrusion detection. Compared with other detection algorithms, Experimental results demonstrated that the proposed KPCA SVM model performed higher predictive accuracy, faster convergence speed and better generalization, implying that the proposed model is successful for intrusion detection.

Find By Topic