THE NEURAL NET APPROACH TO THE HARD TO FORMALIZE TASKS OF ANALYSIS AND PREDICTION

The article discusses the use of neural networks and attempt to reveal the peculiarities of the different types of neural networks in the context of their application for the solution of difficult problems in formalized tasks, as well as how to improve the accuracy of forecasting financial performance, including securities prices. In particular, it discussed in more detail one of these models as a model of so-called self-organizing feature maps of Kohonen.The article discusses the use of neural networks and attempt to reveal the peculiarities of the different types of neural networks in the context of their application for the solution of difficult problems in formalized tasks, as well as how to improve the accuracy of forecasting financial performance, including securities prices. In particular, it discussed in more detail one of these models as a model of so-called self-organizing feature maps of Kohonen