基于XGBoost的肥胖水平綜合預(yù)測(cè)與SHAP模型解釋分析

打開文本圖片集

中圖分類號(hào)：TP399 文獻(xiàn)標(biāo)識(shí)碼：A文章編號(hào)：2096-4706（2025）07-0040-07

Abstract： This paper aims to use the XGBoost model to predict obesity levels and explain the contribution of various featurestoobesityriskthroughtheSHAPmethd，soastoidentifykeyifuencingfactorsandprovideascientificbasisfor obesity prevention.Modeling isconductedbasedonmultiple featuressuchas familyhistoryofobesity，dietaryhabits，and frequencyofphysicalactivityXGBoostisused topredictobesitylevels，andSHAPvaluesareappliedtoanalyzethempactof eachfeatureonthe modeloutput，toexplainthecontributionofeach feature toobesityclasifcation.Family historyofobesity age，and dietary habitsare keyfactors afectingobesity.SHAPanalysis furtherreveals the specificcontributionsandimpactof thesefactorsonobesityclassification.BycombiningtheeffcientpredictiveabilityofXGBoostandtheexplanatoryanalysisof SHAP，thisresearchnotonlyidentifiesthekeyfeaturesthataffctobesitybutalsoprovidesascientificbasisforpersoalized health management and obesity prevention，demonstrating theapplication potential ofMachine Learming inthe fieldof public health.

Keywords： SHAP; XGBoost; Big Data; obesity level; health management

0 引言

對(duì)肥胖水平進(jìn)行數(shù)據(jù)挖掘，是指通過分析與肥胖相關(guān)的各類數(shù)據(jù)，揭示肥胖的成因、發(fā)展趨勢(shì)以及與健康風(fēng)險(xiǎn)之間的關(guān)系，從而為肥胖的預(yù)防、管理和治療提供數(shù)據(jù)支持。（剩余8043字）