基于Borderline-SMOTE和OOA-SVM的心脏病诊断预测模型  

Prediction Model of Heart Disease Diagnosis Based on Borderline-SMOTE and OOA-SVM

作　　者：祖璇 张广海 ZU Xuan;ZHANG Guang-hai(Department of Economics,Wuhu University,Wuhu 241008,Anhui,China;School of Big Data and Artificial Intelligence,Wuhu University,Wuhu 241008,Anhui,China)

机构地区：[1]芜湖学院经济系,安徽芜湖241008 [2]芜湖学院大数据与人工智能系,安徽芜湖241008

出　　处：《兰州文理学院学报(自然科学版)》2025年第1期46-52,共7页Journal of Lanzhou University of Arts and Science(Natural Sciences)

基　　金：安徽高校自然科学研究重点项目(2022AH052900)。

摘　　要：为实现心脏病精准预测,构建了一种预测准确率较高的心脏病诊断预测模型.首先对原始数据集进行pearson相关性分析和归一化处理;然后采用过采样技术Borderline-SMOTE算法,平衡训练数据集的少数类;之后利用鱼鹰优化算法(Osprey Optimization Algorithm,OOA)优化支持向量机(support vector machine,SVM),获得最优参数组合(C,g);最后在测试数据集上进行分类预测.与SSA-SVM、SMA-SVM和SVM相比,本文方法OOA-SVM的预测准确率最高,达到了95.08%,且模型稳定性最好.In order to realize the accurate prediction of heart disease,a prediction model of heart disease diagnosis with high prediction accuracy was established.Firstly,pearson correlation analysis and normalization were performed on the original data set;and then the over-sampling Borderline-SMOTE algorithm was used for balance a few classes of the training data set;then the Osprey Optimization Algorithm(OOA)was used to optimize the support vector machine(SVM)to obtain the optimal parameter combination(C,g);finally,classification prediction is made on the test data set.Compared with SSA-SVM,SMA-SVM and SVM,the prediction accuracy of OOA-SVM is the highest,reaching 95.08%,and the model stability is the best.

关 键 词：Borderline-SMOTE 鱼鹰优化算法 支持向量机 心脏病诊断预测 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]