Methods for Collecting and Classifying Data using a Soundlet Bayesian Neural Network
N. A. Zavalko · International Journal of Advanced Trends in Computer Science and Engineering · 2020
The paper substantiates the application of neural networks for prediction of emergency situations and the stages of building a neural network for simulation of emergency situations.The purpose of the work is achieved by developing scientific and technical bases for the technical implementation of a comprehensive within monitoring, prevention and liquidation system of natural and man-made emergency situations and ensuring environmental safety.Forecasting emergencies and location determination is an urgent task that requires a permanent and effective solution.The basis of this problem is the construction of the effective methods, providing the high speed of the learning pattern recognition models as well as high probability, the adequacy and speed of emergency signals recognition.The article describes the principles of the construction and operation of the system for the recognition of emergency situations using neural networks.Considered apparatus and analysis of the need to use a neural network, to predict the physical parameters the emerging problem.Fulfilled the analysis of the chosen structure and neural networks, which should be used to predict the physical parameters.In article proposed model of a neural network to solve the problem of prediction.Also presented, the mathematical formula for visual understanding of the structure of neural networks and their work.The article is devoted to the problem of emergencies prevention using modern methods of analysis of acoustic data.Emergencies arise in the conditions of extraordinary situations and management in emergency situations characterized by the need for work in the absence of information, the high rate of change in the situation, the need for operational formation of the most effective solutions, which have high efficiency, which imposes requirements to minimize the time and losses in the elimination of emergency situations.Results of the study can be used for analysis and modelling of stability of emergency service that is for evaluation of the emergency or potentially dangerous object.