Device and method for extracting deep belief network information in cognitive information confrontation

The invention provides a deep belief network intelligence extraction device and method in cognitive information confrontation, and relates to the technical field of cognitive electronic warfare. The device comprises a DBN network deep learning platform, a front-end multi-source sensor and a data memory, massive target feature data is collected in real time through the front-end multi-source sensorand input into a DBN network input layer, and original information is uploaded through the input layer; according to the deep learning architecture based on the DBN network, the weight of a generation model is obtained through training of an unsupervised greedy layer-by-layer method, the value is transmitted to a hidden layer through the weight, and on the contrary, the formed weight can be randomly selected to reconstruct original input data; through multi-time iteration of a bidirectional data layer and a hidden layer, obtaining real feature expression of massive reconnaissance data; high-precision analysis intelli