The purpose of the research is to create a secure system for analyzing the state and forecasting processes in the real estate market. Such a system would help realtors, buyers, sellers and all interested parties to choose the necessary residential real estate objects. It is proposed to use an artificial neural network as a model of the relationship between the price of a real estate object and the factors that determine it. The proposed neural network system contains some elements to ensure the security of personal data and other confidential information. To organize the security of the neural network system, an improved model for constructing cryptographic transformations based on the use of two-operand operations is used. The use of such a model allowed the development of a cryptographic transformation method that would provide the necessary high speed of the neural network system for analyzing the state of the real estate market and predicting further scenarios, which guarantees comfortable user work with the system. The correctness of the developed method and models for increasing the speed and stability of group matrix cryptographic transformation was verified. It was proved that the use of the proposed method makes it possible to reduce the complexity of implementing matrix transformation depending on the number of matrix digits from 8 to 33 times.