Efficient Hidden Danger Prediction for Safety Supervision System: An Advanced Neural Network Learning Method

摘要

Aiming at the application of safety supervision hidden danger prediction in education, medical treatment, restaurant, and other tertiary industries, this paper combines ELM algorithm to achieve application of hidden danger prediction. Compared with the traditional BP learning algorithm for NN, the single hidden layer feedforward NN oriented ELM has the advantages of fast learning speed and high generalization performance [8, 11]. The experimental results verify the advantages of the method used in this paper. The method developed here is a novel practice of modern accident-causing theory. In the future, we will combine more effective ELM algorithm to further improve the prediction performance.