%0 Journal Article
%T 基于Smote-XGBoost算法的心脏病预测模型研究<br>A Study of Heart Disease Prediction Model Based on Smote-XGBoost Algorithm
%A 管锦寒
%A 杨健
%A 陈俊钰
%A 李璐
%J Hans Journal of Data Mining
%P 220-234
%@ 2163-1468
%D 2022
%I Hans Publishing
%R 10.12677/HJDM.2022.123003
%X 该模型首先采用合成少数类过采样技术编辑的最近邻来平衡训练数据分布，然后通过集成学习算法XGBoost预测心脏病。为了验证模型效果，本文采用心脏病患者真实医疗数据，利用专家咨询法提取特征，并通过混淆矩阵进行模型评估。与4类基线算法相比，所提模型在AUC、Accuracy、Recall和F-Score指标的评测下均表现良好。实验结果显示，所提模型能够为心脏病预测提供更精准、更智能的辅助参考，同时可以在一定程度上提高诊断的效率和心脏病预测的准确率。<br />
The proposed model uses nearest neighbors edited by synthetic minority class oversampling techniques to balance the training data distribution, and then predicts heart disease by ensemble learning algorithm XGBoost. To detect the prediction reliability, a real medical dataset of heart dis-ease patients are used, features are extracted using expert consultation method, and the model is evaluated by confusion matrix. Compared with the four types of baseline algorithms, the proposed model performs well in terms of AUC, Accuracy, Recall and F-Score metrics. The experimental results show that the proposed model can provide a more accurate and intelligent auxiliary reference for heart disease prediction, and it can also improve the efficiency of diagnosis and the accuracy of heart disease prediction to some extent.
%K 心脏病预测，Smote-Enn算法，XGBoost算法，混淆矩阵，Heart Disease Prediction
%K Smote-Enn Algorithm
%K XGBoost Algorithm
%K Confusion Matrix
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=53188