Intrusion Detection System Of Database Based On User Behavior Of Analysis And Identification

With the rapid development of Internet technology and internet device and the rapid reducing of Internet tariffs, the Internet is used widely. The Internet is changing the models of people’s production and life. But widespread application of the Internet also brings network security issues. The background of knowledge, technical means and collaborative working of intrusion made that the network security issues are becoming the biggest threat to Internet data and resources. Thus, in recent years the network intrusion detection theory and technology are becoming increasingly active. The intrusion detection system of the database based on the analysis of user behavior analysis and recognition and by analysis user behavior dataset of network database. This article reveals the random distribution of network database user behaviors.The main work of the article: explore that BP network improved based on intelligent algorithms applied at the system intrusion detection based on the behavior of Internet users in the database. The algorithm is completed innovative exploration from the following points.1.The slow variability characteristic of biological visual system is introduced to the analysis of user behavior and expression, the noun reveals the slow degeneration of the law of user network behavior.2.The twin-engine warning based on BP neural network and human intervention methods is combined.3.The correlation and redundancy of records is eliminated with ICA algorithm to improve the training effect BP algorithm.4.The dynamic training and updating technology front and back data is used to achieve a self-adaptive BP network.5.This paper proposed a suitable twin-engine intrusion detection mode.6.The C# and Matlab programming is Integrated to improve the speed of development of the project.