Prediction of China’s gold medal count at the Winter Olympics based on machine learning

In recent years, the prediction of Olympic medal counts has remained a popular topic. By exploring factors that influence the trends in Olympic medal counts, predictions can be made regarding these medals. Common influencing factors include national GDP, the number of participating athletes, and the number of events participated in. Therefore, this paper aims to investigate the influencing factors of gold medal counts at the Winter Olympics. Using machine learning algorithms, a predictive model was established by analyzing data such as national GDP, the number of participating athletes, and the number of events participated in. The K-Nearest Neighbors (KNN) regression was applied to model the number of participants, while linear regression was used for the number of events participated in and China’s GDP. These three factors were used as features, comparing different algorithms’ fit to the data, ultimately choosing Random Forest Regression for establishing the gold medal count prediction model. After optimization, the model achieved good performance on the training set with an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of 0.81, MSE of 0.51, RMSE of 0.72, and MAE of 0.41. Through comparisons between actual and predicted values before 2026, the model demonstrated effective results. Finally, experimental validation showed that the model also performed well in predicting the gold medal count for 2026. Gold medal counts are often seen as symbols of national comprehensive strength; a gold medal prediction model helps in formulating differentiated competitive strategies and enhancing international influence. By foreseeing future gold medal outcomes in advance, analyzing influencing factors, and making targeted improvements, the probability of winning gold medals can be increased.