Network Intrusion Detection Method Based on Relevance Deep Learning

With the development of science and technology, computer network has been more and more widely used, but its inherent characteristics cause that it is prone to various invasions. Its security research is very valuable. As an active security mechanism, intrusion detection technology is the key to ensure network security. This paper proposed an network intrusion detection method based on the relevance deep learning, which learned the learning principle of relevance deep and the training algorithm of restricted Boltzmann machine, which analyzed the principle of feasibility that the relevance deep learning was applied to the network intrusion detection system in the network intrusion detection technology. The relevance depth learning was applied to the network intrusion detection technology, which could obtain the higher detection accuracy. The simulation results show that the network intrusion detection method based on relevance deep learning has a high level of average detection rate and average false detection rate for unknown intrusion and attack. The experiment results show that the proposed method is reliable and effective.