As the traditional Bayesian network intrusion detection technology exists some problems of low detection efficiency owing to too much calculation and the single data source resulting in a certain influence on the accuracy of the examination, puts forward a hybrid Bayesian intrusion detection technology based on factor analysis. The method applies factor analysis to the attribute features selection of network connecting data and reduces the data correlation, at the same time, integrates network data and the host data for analysis and evaluation of the current network security state to improve the detection accuracy. The experimental results show that the proposed detection technology greatly reduces the dimension of data and improves the computational efficiency and detection accuracy.