Training forecast to football athletes using Hopfield neural networks based on Markov matrix.

Abstract

This paper proposes a neural network based on the Markov probability transition matrix to predict the training performance of football athletes. Firstly, seven training indicators affecting the training performance are designed by the Event-group training theory. Then, a discrete Hopfield neural network is employed according to the seven training indicators. To improve the forecast ability of the discrete Hopfield neural network, the Markov probability transition matrix is used to calculate the activation probability of neurons. Finally, experimental results indicate that the proposed model defeats against the competitors in the forecast of training performance of football athletes. And the proposed model can find the major training indicators that have direct effects on the training performance, which can provide scientific suggestions for coaches to customize training plans. We demonstrate that the seven training indicators can sufficiently evaluate the effectiveness of training plans in the improvement in terms of training performance for football athletes.