Scholarlink

<p>This study leverages machine learning techniques to explore the socio-demographic determinants that influence sports habits among the Italian population. Through the use of data from the ‘Aspects of Daily Life’ survey conducted by the Italian National Institute of Statistics, with reference to the latest available year, 2021, this work investigates how different socio-demographic factors, such as age, gender, education, marital status and regional disparities, affect individuals’ participation in sports activities. By employing an eXtreme Gradient Boosting (XGBoost) model, renowned for its predictive accuracy and efficiency, this work identifies the most significant predictors of sports habits. The interpretation of the model is further enriched by SHAP (SHapley Additive exPlanations) values, thus providing a detailed understanding of the impact of each socio-demographic variable on the likelihood of engaging in regular sport activities. This approach not only highlights the critical factors but also underscores the potential for implementing targeted interventions to promote physical activity across various demographic groups in Italy.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Harnessing machine learning to investigate the socio-demographic determinants of sports habits in Italy

  • 期刊论文

  • 发表日期：

    2025年4月29日

查看全文
  • Fabrizio Antolini,
  • Samuele Cesarini,
  • Ivan Terraglia

摘要

This study leverages machine learning techniques to explore the socio-demographic determinants that influence sports habits among the Italian population. Through the use of data from the ‘Aspects of Daily Life’ survey conducted by the Italian National Institute of Statistics, with reference to the latest available year, 2021, this work investigates how different socio-demographic factors, such as age, gender, education, marital status and regional disparities, affect individuals’ participation in sports activities. By employing an eXtreme Gradient Boosting (XGBoost) model, renowned for its predictive accuracy and efficiency, this work identifies the most significant predictors of sports habits. The interpretation of the model is further enriched by SHAP (SHapley Additive exPlanations) values, thus providing a detailed understanding of the impact of each socio-demographic variable on the likelihood of engaging in regular sport activities. This approach not only highlights the critical factors but also underscores the potential for implementing targeted interventions to promote physical activity across various demographic groups in Italy.