Research on Momentum Quantification Based on Genetic Algorithm Optimising Neural Networks

Firstly, the study examines the required variables with Pearson's correlation coefficient and constructs a capture model. Then, a genetic algorithm was used to optimize the neural network to optimize the weight parameters, quantify the momentum influence parameters and construct a capture model to identify the players' tables at different moments to provide a visual description of the game flow. Then, the reliability of the capture model was verified using linear regression test with one time exponential smoothing method for regression analysis to show that the player's swing and success in the game is not random and time series model was used for prediction analysis. Finally, based on the multi-indicator evaluation technique of principal component analysis, the model is established to predict the game swings, considering the momentum fluctuations, which can point out the irrationality among the techniques of both the players themselves and their opponents as well as their strengths, so as to enable them to better target their training. Provide technical theoretical support for players in the field of sports competition.