A multi-class classification MCLP model with particle swarm optimization for network intrusion detection.

The critical data we share through computer network gets stolen by unethical means. This unethical way of accessing one's data without proper authentication becomes intrusion. To solve this issue, in this paper we propose a new network intrusion detection method, Multi-Class Classification Multiple Criteria Linear Programming (MCC-MCLP) model. MCLP is a mathematical classification technique that is used widely to solve real-time data mining problems. So far, the literature discusses only about binary classification MCLP. But in this paper we propose a Multi-Class Classification MCLP model. We use PSO for fine-tuning the parameters of MCC-MCLP. KDD CUP 99 data set is used for performance evaluation of the proposed method. Our MCC-MCLP method classifies the data better and helps in fine-tuning the parameters with the help of PSO. The results clearly show that the proposed model performs better in terms of detection rate, false alarm rate and accuracy. [ABSTRACT FROM AUTHOR]