Prediction of sports strength based on linear regression and random forest model
DOI:
https://doi.org/10.62051/rgn3nx51Keywords:
Linear Regression; Random Forest Model; Dynamic Weighted Model; Exponential Decay Coefficient.Abstract
This paper constructs a dynamic weighted comprehensive model that integrates linear regression and random forest algorithms. This model is used to predict the distribution of competitive performance indicators in international comprehensive sports events and analyze the relevant influencing factors. The model innovatively integrates multiple factors. It incorporates historical data and uses a mechanism to assign higher weights to recent performances, so as to better reflect the current situation. The model also takes into account geographical advantages, which may affect a country's performance in certain sports. Additionally, it includes the effectiveness of training strategy decision - making entities and the scale of participants. In the quantification process, the model first uses linear regression to predict the number of entities that are likely to achieve the best results for the first time. Then, it quantifies the competitiveness index of each project. Specifically, competitive achievements are converted into scores, and a dynamic weighting method is adopted to adjust the scores according to the time when the achievements are obtained, with more emphasis on recent achievements. Finally, the random forest algorithm is used to comprehensively consider various factors and predict the final results of participating units. This model has high application feasibility. Its unique dynamic weighting mechanism and multi - factor integration make its predictions more accurate. The model also predicts potential new - achiever units and evaluates prediction uncertainty. Future research could incorporate real - time individual data and socio-economic variables to enhance the model.
Downloads
References
[1] Maulud D, Abdulazeez A M. A review on linear regression comprehensive in machine learning[J]. Journal of applied science and technology trends, 2020, 1(2): 140-147. DOI: https://doi.org/10.38094/jastt1457
[2] Martin P. Linear regression: An introduction to statistical models[J]. 2022. DOI: https://doi.org/10.4135/9781529682731
[3] Li Y, Mu Y. Research and performance analysis of random forest-based feature selection algorithm in sports effectiveness evaluation[J]. Scientific Reports, 2024, 14(1): 26275. DOI: https://doi.org/10.1038/s41598-024-76706-1
[4] Gao Z, Kowalczyk A. Random forest model identifies serve strength as a key predictor of tennis match outcome[J]. Journal of Sports Analytics, 2021, 7(4): 255-262. DOI: https://doi.org/10.3233/JSA-200515
[5] Fernando K R M, Tsokos C P. Dynamically weighted balanced loss: class imbalanced learning and confidence calibration of deep neural networks[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 33(7): 2940-2951. DOI: https://doi.org/10.1109/TNNLS.2020.3047335
[6] Ahmad S, Ullah A, Akgül A, et al. A Novel Homotopy Perturbation Method with Applications to Nonlinear Fractional Order KdV and Burger Equation with Exponential‐Decay Kernel[J]. Journal of Function Spaces, 2021, 2021(1): 8770488. DOI: https://doi.org/10.1155/2021/8770488
[7] Christoph S, L. S S, Dominik S, et al. Forecasting the Olympic medal distribution – A socioeconomic machine learning model[J]. Technological Forecasting & Social Change,2022,175Christoph A H, K A B, Bergita G. Learning from machine learning: prediction of age-related athletic performance decline trajectories. GeroScience, 2021, 43(5): 1–13.
[8] Christoph A H, K A B, Bergita G. Learning from machine learning: prediction of age-related athletic performance decline trajectories. [J]. GeroScience,2021,43(5):1-13. DOI: https://doi.org/10.1007/s11357-021-00411-4
[9] Zhonghui T, Juan H, Shuo M, et al. Estimating cloud base height from Himawari-8 based on a random forest algorithm[J]. International Journal of Remote Sensing,2020,42(7):2485-2501. DOI: https://doi.org/10.1080/01431161.2020.1854891
[10] Jiandong H ,Tianhong D ,Yi Z , et al. Predicting the Permeability of Pervious Concrete Based on the Beetle Antennae Search Algorithm and Random Forest Model[J]. Advances in Civil Engineering, 2020. DOI: https://doi.org/10.1155/2020/8863181
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Transactions on Computer Science and Intelligent Systems Research
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
