Radial Neural Network Processing Applied to Athlete’s Personalized Psychological Regulation Detection System

Abstract

Radial neural network is based in the world with the characteristics of active adaptation, active learning, active recognition, low error rate, and thought mapping and plays an important role in personalized regulation. However, in practical applications, due to the existence of pattern recognition, motion control, and a large amount of combined knowledge, the traditional methods are difficult to solve, even powerless, and cannot be effectively solved. Although the traditional BP network is more widely used, the BP neural network is easy to enter the regional minimum value during the training process, which leads to a lower training learning speed and low efficiency. The RBF network (radial neural network) is in a certain sense, and it can detect both known intrusions and unpredictable intrusions. At the same time, it is superior to BP neural network in data collection, pattern recognition, and personality customization. Through detection and comparison, it is found that the radial network has improved the analysis speed by about 20%, and the degree of privacy protection of athletes is as high as 99%, which is close to the full value, and the accuracy of the psychological control scheme is also improved by about 15% on the basis of the classic network. Athletes can be adjusted as soon as possible to achieve the best state; so, it will be an ideal choice to use a radial neural network for the psychological adjustment detection system.