PCA-based BP neural network detection system, method and application for information security

The invention discloses a PCA-based BP neural network detection system, method and application for information security. The PCA-based BP neural network detection system comprises the steps of collecting data of risk factors influencing information security recently and simulating experts to evaluate an evaluation factor set; constructing a training sample and a prediction sample; carrying out dimension reduction processing on the data in the training sample and the prediction sample by utilizing SVD; performing fuzzy cerebellar neural network training by using the training sample after dimension reduction; and substituting the dimension-reduced prediction sample into the trained fuzzy cerebellar neural network to obtain a prediction result. Meanwhile, the invention discloses a corresponding system and application. According to the method, dimension reduction, decomposition and neural network training are sequentially performed on the data based on risk factor types and risk level data which influence informa