基于集成学习融合模型的血糖预测

Blood Glucose Prediction Based on Ensemble Learning Fusion Model

介绍集成学习预测方法,阐述集成学习在血糖预测中的应用,基于个体常规体检数据,使用集成学习的方法,融合线性回归、梯度提升决策树、随机森林等模型对血糖进行预测,实验结果表明该方法对血糖具有更高的预测精度并能更准确地识别血糖异常个体。

The paper introduces the ensemble learning prediction method and dilates on the application of ensemble learning in blood glucose prediction.Based on individual routine physical examination data,it predicts blood glucose through the ensemble learning method that is combined with linear regression,gradient boosted decision tree,random forest and other models.The experimental results indicate that the method boasts higher prediction precision for blood glucose and is able to identify individuals with abnormal blood glucose more accurately.