Analysis of Key Components in the Propagation of Artificial Neural Networks
Abstract
Today, the use of neural network technologies in solving practical problems in various fields of science and technology has become a topical issue. Such problems that can be successfully solved by artificial neural networks include: - forecasting emergency situations in hazardous production facilities, classification, diagnosis and identification of emergency situations. Since the creation of the first artificial neural networks, they have been used both to solve various practical problems and to study possible rules of brain activity. Neural networks are successfully used in a wide range of applications such as pattern recognition, forecasting, data compression, control tasks, etc. Neural network models claim to be a universal device that solves various specific problems from different problem areas of GIS. This versatility is due to the fact that neural network technologies provide a standard way to solve many problems. Integration of tools based on neural network technologies non-standard tasks. Solving poorly formalized problems and geographic information systems significantly increases the quality and speed of information processing, expands their capabilities to perform practical, scientific, educational and other tasks. This article discusses other basic artificial neural network concepts that are very common in the practice of neural network analysis and modeling.