Research on Momentum prediction algorithm for athletes based on the CNN-LSTM Model

Ziyi Xu; Qiya Shu; Yilan Ye

doi:10.62051/mx0t8r76

Authors

Ziyi Xu
Qiya Shu
Yilan Ye

DOI:

https://doi.org/10.62051/mx0t8r76

Keywords:

Sports events, Momentum, Prediction Model, CNN-LSTM.

Abstract

Momentum has been widely recognized as a psychological factor affecting athletes' performance during sports events. Quantifying the momentum of both athletes in a match and predicting its future trends and reversals can provide effective information for coaches to make tactical decisions during a match. In this paper, taking the Wimbledon dataset as an example, the quantitative model of momentum is firstly constructed by the rank-sum-ratio evaluation method based on EWMA. Subsequently, this paper introduces the CNN network to process the data input based on the traditional LSTM network, takes the output data of the CNN network as the input layer of the LSTM network, and constructs a CNN-LSTM model to predict the change of momentum of athletes in the match, and the reversal time of momentum of athletes of both sides points. Through the model comparison analysis, this paper found that the prediction accuracy of the CNN-LSTM in different time periods is up to 85%, surpassing the LSTM network by 10%. This paper helps coaches and athletes to identify the future trend of momentum in the course of the match so that they can make reasonable tactical arrangements to achieve better performance in sports events.

Downloads

Download data is not yet available.

References

[1] Briki W. Rethinking the relationship between momentum and sport performance: Toward an integrative perspective[J]. Psychology of Sport and Exercise, 2017, 30: 38-44.

[2] Briki W, Den Hartigh R J R, Markman K D, et al. How psychological momentum changes in athletes during a sport competition[J]. Psychology of Sport and Exercise, 2013, 14(3): 389-396.

[3] Iso-Ahola, Seppo E., and Charles O. Dotson. Psychological momentum: Why success breeds success[J]. Review of general psychology, 2014, 18(1): 19-33.

[4] Xu J. Prediction and planning of sports competition based on deep neural network[J]. Computational Intelligence and Neuroscience, 2022, 2022.

[5] Bunker R P, Thabtah F. A machine learning framework for sport result prediction[J]. Applied computing and informatics, 2019, 15(1): 27-33.

[6] Yu Y, Si X, Hu C, et al. A review of recurrent neural networks: LSTM cells and network architectures[J]. Neural computation, 2019, 31(7): 1235-1270.

[7] Li C, Ji Y, Liu J, et al. Momentum Study Based on Logistic Regression Models and BP Neural Networks[J]. Journal of Computing and Electronic Information Management, 2024, 12(2): 115-119.

[8] Lin J, Shao P, Zhang Q. Advancing Tennis Analytics: Comprehensive Modeling for Momentum Identification and Strategic Insights[J]. International Journal of Computer Science and Information Technology, 2024, 2(1): 104-117.

[9] Sunny M A I, Maswood M M S, Alharbi A G. Deep learning-based stock price prediction using LSTM and bi-directional LSTM model[C]//2020 2nd novel intelligent and leading emerging sciences conference (NILES). IEEE, 2020: 87-92.

[10] Livieris I E, Pintelas E, Pintelas P. A CNN–LSTM model for gold price time-series forecasting[J]. Neural computing and applications, 2020, 32: 17351-17360.

[11] Karsoliya S. Approximating number of hidden layer neurons in multiple hidden layer BPNN architecture[J]. International Journal of Engineering Trends and Technology, 2012, 3(6): 714-717.

[12] Goutte C, Gaussier E. A probabilistic interpretation of precision, recall and F-score, with implication for evaluation[C]//European conference on information retrieval. Berlin, Heidelberg: Springer Berlin Heidelberg, 2005: 345-359.

[13] Yu G, Wang X, Dong B. Revealing the Close Connection Between Tennis Match Momentum and Victory Through Machine Learning Methods[J]. Highlights in Science, Engineering and Technology, 2024, 98: 287-294.

[14] Hua H, Sun W. Evaluating Momentum-Weighted LSTM Models for Predicting Tennis Match Outcomes[J]. Highlights in Science, Engineering and Technology, 2024, 101: 704-711.