Author(s)

TiLiang Zhang, JunJie Chen, Cheng Cheng, Xing Li^*

Affiliation(s)

School of Mathematics and Statistics, Hubei University of Education, Wuhan 430205, Hubei, China.

Corresponding Author

Xing Li

ABSTRACT

Amid the expanding scale of the Olympic Games, precise forecasts of the event portfolio and medal allocation have become critical for national strategic planning. Olympic data, however, are markedly non-linear and structurally dynamic, rendering traditional linear methods inadequate. This paper therefore develops an integrated forecasting framework that estimates discipline-level event counts and country-specific medal shares for the 2028 Games. Athletes were first aggregated by team and country, and analyses were conducted at the discipline level to prevent overgeneralisation inherent in sport-level aggregation. Support Vector Regression was employed to model the relationship between historical covariates and the number of events per discipline; the resulting predictions achieved a mean-squared error of 1.744 and an R2 of 0.634. The strategic salience of each discipline to individual nations was subsequently quantified via weighted medal totals and visualised through rose plots. Medal shares were derived by mapping historical performance indicators to fractional medal outcomes using XGBoost, after an initial recurrent architecture exhibited convergence difficulties. These fractions were scaled by the projected event counts, and a calibrated 15 % host-nation uplift was applied to the United States before global normalisation. The resulting projection allocates 47, 45 and 36 medals to the United States, 35, 24 and 15 to China, and 18, 9 and 10 to Japan. Retrospective validation against 2024 data places all nine reference nations within 95 % prediction intervals, confirming the framework’s reliability. This study can provide data support for national sports management departments and optimize the allocation of training resources.

KEYWORDS

Discipline-level events; XGBoost; Olympic forecasting; Support Vector Regression; Resource allocation

CITE THIS PAPER

TiLiang Zhang, JunJie Chen, Cheng Cheng, Xing Li. Nonlinear prediction based on 2028 Olympic events and medals. World Journal of Information Technology. 2025, 3(4): 51-57. DOI: https://doi.org/10.61784/wjit3053.

REFERENCES

[1] Bo Z, Chaoling Q, Xiaoli X, et al. GM (1, 1) Model Gray Prediction for the Gold-Medal Result of Women's Put Shot in the 30th Olympic Games. 2011 International Conference on Future Computer Science and Education, Xi'an, China. IEEE, 2011, 334-337.

[2] Moolchandani J, Chole V, Sahu S, et al. Predictive Analytics in Sports: Using Machine Learning to Forecast Outcomes and Medal Tally Trends at the 2024 Summer Olympics. 2024 4th International Conference on Technological Advancements in Computational Sciences (ICTACS), Tashkent, Uzbekistan. IEEE, 2024, 1987-1992. DOI: 10.1109/ICTACS62700.2024.10840553.

[3] Sagala N T M, Ibrahim M A. A Comparative Study of Different Boosting Algorithms for Predicting Olympic Medal. 2022 IEEE 8th International Conference on Computing, Engineering and Design (ICCED), Sukabumi, Indonesia. IEEE, 2022, 1-4. DOI: 10.1109/ICCED56140.2022.10010351.

[4] Wang Shiyu. Olympic medal prediction model based on nonlinear regression and BP neural network. Sports Goods and Technology. 2017(24): 4-5+83.

[5] Dong Qi, Gao Feng. Using Support Vector Machine Method to Predict the Number of Chinese Medals at the 2016 Rio Olympics. Sports. 2016(03): 1-4.

[6] Yan Yuyang. Olympic medal prediction based on grey theory. Journal of Sichuan University of Arts and Sciences, 2011, 21(05): 21-23.

[7] Luo Yubo, Cheng Yanfang, Li Mengyao, et al. Prediction of China's medal count and overall strength for the Beijing Winter Olympics: based on the host effect and grey prediction model. Contemporary Sports Technology, 2022, 12(21): 183-186.

[8] Yan D. Olympic Model Perdiction and Analysis based on LSTM and Topsis Models. Journal of Computer Science and Electrical Engineering, 2025, 7(3): 1-10.

[9] Cheng Hongren, Lv Jie, Yuan Tinggang. Prediction of China's Track and Field Performance at the Tokyo Olympics from the 2018 World Top 20 Athletics Rankings. Sports Science and Technology Literature Bulletin, 2020, 28(04): 4-8.

[10] Shi Huimin, Zhang Dongying, Zhang Yonghui. Can Olympic medals be predicted? ——From the perspective of interpretable machine learning. Journal of Shanghai Sport University, 2024, 48(04): 26-36.