Taboo and artificial bee colony two-way optimized support vector machine-based intrusion detection method

Abstract

The invention discloses a taboo and artificial bee colony two-way optimized support vector machine-based intrusion detection method. Firstly an original artificial bee colony algorithm is reconstructed; an overall thought is that a honey source and to-be-optimized parameters of a support vector machine are synchronously coded; an initial solution of the honey source is generated in a random generation mode; and a taboo-artificial bee colony-based search strategy is executed to search for an optimal feature subset. The method specifically comprises the steps of introducing a first taboo table in an initial employed bee search stage, performing storage memory on found local optimal solutions, and avoiding the local optimal solutions in subsequent predetermined-time iteration; introducing a second taboo table in an observation bee stage, and introducing a reward and punishment mechanism to improve a fitness function of the honey source, thereby improving the diversity of the solutions; and finally introducing a third taboo table in an investigation bee stage to store the solutions with maximum development frequencies and unimproved fitness values. The method has relatively high feasibility and high practicality.

Description

[0001] technology neighborhood [0002] The invention relates to the technical field of network security, in particular to an intrusion detection method based on a taboo and artificial bee colony bidirectional optimization support vector machine. Background technique [0003] In recent years, with the increasing scale of the network, while the computer network has brought convenience to people's life and work, the methods of network intrusion and attack have become more complex, and the losses caused by various destructive network attacks have become increasingly serious. Although traditional security defense methods and strategies have played a protective role in network security to a certain extent, they have been unable to prevent complex, changeable and increasingly rampant intrusions. Intrusion Detection System) came into being. This kind of active protection strategy technology has been valued by experts and scholars at home and abroad, and how to study an effective int...

AI Technical Summary

Problems solved by technology

[0005] What the present invention aims to solve is that the feature selection search efficiency in the existing intrusion detection model is not high, the minimum feature subset search cannot be solved, and the search and modeling are difficult to optimize parameters simultaneously. Intrusion Detection Method Based on Group Bidirectional Optimal Support Vector Machine

Images