Title :
Intrusion Detection Based on Minimax Probability Machine with Immune Clonal Feature Optimized
Author :
Chen, Zhenguo ; Li, Dongyan ; Ren, Hongde
Author_Institution :
Dept. of Comput. Sci. & Technol., North China Inst. of Sci. & Technol., Beijing
Abstract :
This paper synthetically applied immune clonal feature optimized method and minimax probability machine to the intrusion detection. Minimax probability machines (MPMs) is a state-of-the-art classification algorithm and has higher performance than traditional methods. We use fusions of immune clonal algorithm and MPM to enhance the overall performance of MPM. Immune clonal algorithm is used to optimize the feature so as to generate newly features to boost MPMs. So a new classifier model based on MPMs with immune clonal feature optimized is proposed and is applied to intrusion detection. The dataset kddcup99 is our experiment data, the experimental results show that the new MPMs with immune clonal feature optimized can give higher recognition accuracy and need less training time than the general MPM.
Keywords :
minimax techniques; probability; security of data; immune clonal algorithm; immune clonal feature optimization; intrusion detection; minimax probability machine; state-of-the-art classification algorithm; Classification algorithms; Covariance matrix; Immune system; Intrusion detection; Machine learning; Minimax techniques; Optimization methods; Protection; Support vector machine classification; Support vector machines; Data Classification; Immune clonal; Intrusion Detection; Minimax Probability Machine; Network security;
Conference_Titel :
Knowledge Acquisition and Modeling, 2008. KAM '08. International Symposium on
Conference_Location :
Wuhan
Print_ISBN :
978-0-7695-3488-6
DOI :
10.1109/KAM.2008.86