Analysis of Influencing Factors and Prediction of Kinetic Energy of Tennis Based on Factor Analysis and LSTM Model

By analyzing the competitive data of the players in the tennis match, this paper evaluates the players' performance by using factor analysis method, analyzes some key factors that affect the players' competitive state, and uses LSTM to predict the situation change in the match. Ten factors are extracted from the data set, and the factor analysis model is used to reduce the dimension of the data, and four main factors are identified. Based on this, the factor scoring model is established to obtain the comprehensive score and performance of players. Descriptive statistics and normality test were carried out for related factors. Mann-Whitney U test was used to explore the correlation between player performance and score. It follows that there are significant differences in the scores of these five factors, and there is a direct correlation between a player's “motivation” and his performance in the game. The “momentum” of players is easily affected by every indicator in the match. Nine indicators are taken as the “momentum” factors affecting tennis players, and the LSTM prediction model is established to predict the change of “momentum” in the match. The test results show that LSTM has good predictive performance. In addition, the influence factors of tennis players' “momentum” are analyzed by using decision tree, and it is found that the two factors that have the greatest influence on players' “momentum” are service error and successful break. Given the above analysis, in a new round of matches, players should use the success of breaking serves and other ACES to increase their “momentum” on the court. Quantifying and visualizing player performance to accurately reflect changing circumstances during the game. In addition, by adding other relevant factors, the generalization of the prediction model is improved, so that it has a good performance in different types of competitive sports.