A New Intrusion Detection Method Based on BPSO-SVM

• Published in: 2008 International Symposium on Computational Intelligence and Design Vol. 1; pp. 473 - 477
• Main Authors: Jing Ma, Xingwei Liu, Sijia Liu
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-10-2008
• Subjects: Algorithm design and analysis; Binary PSO; Computational intelligence; Computer networks; Design engineering; Feature Selection; Intrusion detection; Machine intelligence; Mathematics; Parameters Optimization; Particle swarm optimization; Support vector machine classification; Support vector machines; SVM
• ISBN: 0769533116; 9780769533117
• DOI: 10.1109/ISCID.2008.65

Intelligent algorithms being applied in intrusion detection system (IDS) becomes a tendency in recent years. This paper presents a new method of hybrid detection based on BPSO-SVM, a mixed algorithm that is composed of modified binary particle swarm optimization (BPSO) and support vector machine (SVM). This algorithm proposes a simultaneous feature selection and SVM parameters optimization. Experiments on KDD CUP'99 dataset show that this method can be an effective way for hybrid detection.