【摘要】 我国心血管病患病率和死亡率上升，高血压患者达2.45亿。制定有效方法控制血压，改变生活习惯对此至关重要且具有较高效益。为筛选出对血压影响较为重要的因素，对比了传统中位数回归、XGBoost和XGBoost中位数回归。XGBoost中位数回归应用较传统方法更优，适用非线性复杂数据处理。通过构建XGBoost中位数回归模型得出对老年人群而言，适宜的参加社交活动和良好的睡眠都有助于高血压控制。而身体残疾如失明和肢体残障增加了高血压的治疗难度。为残疾患者制定个性化的计划可帮助控制血压，提高生活质量。更多还原

【Abstract】 The prevalence and mortality of cardiovascular diseases have increased in China, and the number of hypertension patients has reached 245 million. Developing effective methods to control blood pressure and changing lifestyle habits are crucial and highly beneficial.In order to identify factors significantly impacting blood pressure, the traditional median regression, XGBoost and XGBoost median regression are compared. XGBoost median regression is proved superior to traditional methods for handling nonlinear complex data. The establishment of XGBoost median regression model shows that engaging in social activities and holding good sleep contribute to blood pressure control for the elderly. However, physical disabilities, such as blindness and limb impairments, increase the difficulty of managing hypertension. Making personalized plans for disabled patients can help control blood pressure and enhance their life quality.更多还原