DOI: 10.66106/tiyua5.20250106 ISSN: 3105-7608

基于随机森林和灰色GM（1,1）对四人雪车成绩分析及米兰冬奥会混合预测研究（Performance Analysis of Four-man Bobsleigh and Hybrid Prediction for Milan Winter Olympics Based on Random Forest and Grey GM(1,1) Model）

徐嘉欣 Jiaxin Xu, 尹一全 Yiquan Yin
Show PDF Cite
Abstract：To provide a theoretical reference for improving the competitive performance of China's four-man bobsleigh team, this study proposes a hybrid prediction method integrating random forest regression and the grey GM(1,1) model. Based on the single-run time data of the top three athletes in the four Winter Olympics from 2010 to 2022, combined with track physical parameters, a track difficulty coefficient and a weighted index were constructed. The random forest model was used to quantify the influence of track characteristics on performance, while the grey model captured long-term performance trends. The results show that vertical drop and number of curves were the most significant factors affecting performance. The grey model achieved a mean relative error of 2.88%, a relational degree of 0.833, and a variance ratio of 0.307, meeting the first-class accuracy standard. The predicted optimal run times for the top three in the 2026 Milan Winter Olympics are 58.66 s, 60.72 s, and 60.99 s, respectively. This method effectively addresses the challenge of predicting performance in winter sports with small sample sizes, providing a scientific basis for China's bobsleigh preparation and a methodological reference for similar ice and snow racing sports.

More from our Archive