Abstract
In recent years, with the rapid development of information and network technology, computer and network infrastructure has become a popular target of hacker attacks. The intense demand for electronic commerce has intensified the growth of hacking incidents. Network security is a systematic concept, and the effective security policy or scheme is the primary goal of network information security. So, the intrusion detection is one of the main research directions in the field of network security. Because the intrusion detection system(IDS) has its own limitations and technical difficulties, how to apply all kinds of intelligent artificial intelligence algorithms to intrusion detection technology is the key to improve the efficiency of intrusion detection. In order to solve the problem of traditional intrusion detection algorithm in the presence of high false negative rate and high false positive rate, combined with the advantages of BP neural network algorithm, this paper puts forward a kind of intrusion detection algorithm which is used the genetic algorithm to optimize the BP neural network algorithm. Firstly, we find the most suitable weights of BP neural network by genetic algorithm. Then, we use the optimized BP neural network for model learning and testing. Simulation results show that compared with the traditional network intrusion detection algorithm, the training time is shorter, and the algorithm has better recognition rate and detection rate.
