决策树模型在2型糖尿病患者脑梗死风险预测中的应用

Application of decision tree C5.0 in type 2 diabetic′s risk prediction of cerebral infarction

目的 探讨决策树在预测2型糖尿病患者发生脑梗死风险中的应用,为进一步的预防和治疗提供参考依据.方法 采用C5.0决策树算法建立预测模型,并对其预测准确性进行评估,全部分析在SPSS Clementine12.0和SPSS13.0软件中完成.结果 所建立的预测模型对训练样本和测试样本的预测准确率分别为88.41%、85.00%,准确率较高.结论 采用C5.0决策树算法预测2型糖尿病患者发生脑梗死的风险是可行的,其预测结果可以为患者自我保健和医务人员制订治疗方案提供依据.

Objective To discuss the application of decision tree C5.0 in type 2 diabetic＇s risk prediction of cerebral infarction in order to provide reference for further prevention and treatment. Methods C5.0 decision tree algorithm was adopted to establish a prediction model and the accuracy of the prediction was assessed. All of the analyses were completed in Clementine12.0 and SPSS13.0. Results The accuracy rates of the model＇s prediction for training sample and testing sample were 88.41% and 85% respectively. The accuracy was high. Conclusion It is practicable to predict type 2 diabetic＇s risk prediction of cerebral infarction with C5.0 decision tree algorithm. The prediction can provide reference for patients＇ self-care and for medical staff to formulate treatments.