Risk early warning safety model for sports events based on back propagation neural network machine learning

In order to ensure the smooth progress of sports events and reduce the possibility of risk accidents, a risk early warning safety model of sports events based on BP neural network and fuzzy theory is established, which has great advantages. The establishment of the model is divided into five steps: (1) using BP neural network as the basic network structure to obtain the fuzzy data; (2) using fuzzy theory to form the eigenvalue matrix; (3) optimizing the evaluation of the early waiting warning data; (4) determining the model; (5) determining the comprehensive numerical value of sports events risk according to the weight and threshold range of the obtained fuzzy optimized BP neural network, alarm based on relevant indicators. In the empirical analysis, the appropriate sample data was selected, the BP neural network topology structure was construct, the simulation training was carried out, the training error results of the input neural network nodes were calculated. Bying comparing the output data of the neural network with the theoretical data it is found that this risk early warning safety model has certain validity and reliability, and can achieve better early warning effect.