The Intrusion Detection Model based on Parallel Multi - Artificial Bee Colony and Support Vector Machine

Abstract

In view of the problems existing in feature selection and support vector machine model parameter optimization in network intrusion detection, artificial bee colony algorithm is introduced. For the artificial bee colony algorithm, there are problems such as easy precocity, poor diversity of the solution, easy to fall into local optimum, and slow convergence in the later stage. In order to relieve these problems, we redesign the algorithm, including honey source coding scheme, the initialization of population, the construction of the fitness evaluation function, the neighborhood search method and so on. Then we propose the synchronization optimization model of characteristic parameters. It overcomes the above defects of the classical ABC algorithm. Finally, we propose an intrusion detection model based on the improved artificial bee colony algorithm and support vector machine model. The experimental results show that the detection performance of our model is far superior to the methods based on other feature selection and detection principles.