MEDAL PREDICTION FOR THE LOS ANGELES OLYMPIC GAMES BASED ON BAYESIAN OPTIMIZED BOOST REGRESSION

To support resource allocation and strategic preparation for the 2028 Olympic Games, this study introduces a Bayesian-optimized Boost framework. First, key variables are extracted through feature engineering, including economic indicators such as GDP, population, host country effects, and recent rule changes in events. Next, Bayesian optimization is used to fine-tune the model's hyperparameters, and multiple metrics like Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and R-squared (R2) are employed to evaluate the model's performance. Then, time-series cross-validation based on 2024 data shows that the model achieves an R2 of 0.91, outperforming baseline models such as decision trees and standard Boost. Predictions indicate that 23 delegations will gain more medals, while 46 countries may see a decline. Finally, k-means clustering is used to identify each country's dominant sports, quantify the impact of the host country effect on event selection, and provide data support for preparation.