SUMMARY
The article deals with model evaluation system level of data protection with the use of neural networks . The aggregate of security, which correspond to the object of information activities identified technical channels of information leakage at a given time can be represented as dynamical systems. Environment for neural network system ( NNS ) assess the level of data protection can be represented as a set of discrete – event systems with discrete bound states of technological security. The structure of neural classifier that implements the decision function estimation of information security.