| 摘 要: 针对现有的心脏病风险预测模型普遍存在准确率低和可解释性差的问题,提出了一种基于改进鲸鱼优化算法和集成技术的心脏病预测模型。首先,以随机森林为基础分类器与逻辑回归模型进行集成;其次,改进鲸鱼优化算法,使用改进后的算法调整模型的最优超参数;最后,评估模型并使用沙普利加性解释方法进行可解释性分析。使用多个公开数据集进行实验,结果表明该模型达到了96.8%的准确率,精确率为96.2%,召回率为97.7%,F1分数值为96.9%,AUC值为0.989,其效果显著强于其他机器学习模型。 |
| 关键词: 机器学习 改进鲸鱼优化算法 集成技术 心脏病预测 可解释性 |
|
中图分类号: TP391
文献标识码: A
|
|
| Interpretable Heart Disease Prediction Model Based on an Improved Whale Optimization Algorithm |
|
ZHANG Long, WANG Xiaoxia, LI Xiang, CHEN Xiao
|
(School of Electronic Information and Artificial Intelligence, Shaanxi University of Science & Technology, Xi’an 710021, China)
zl1342145664@163.com; wangxiaoxia@sust.edu.cn; lixiangzj@sust.zdu.cn; chenxiao@sust.edu.cn
|
| Abstract: To address the common issues of low accuracy and poor interpretability in existing heart disease risk prediction models, this paper proposes a heart disease prediction model based on an improved Whale Optimization Algorithm and ensemble techniques. Firstly, an ensemble model is constructed using Random Forest as the base classifier and a Logistic Regression model. Subsequently, the standard Whale Optimization Algorithm is improved, and the enhanced algorithm is employed to fine-tune the model’s optimal hyperparameters. Finally, the model is evaluated, and interpretability analysis is conducted using the Shapley Additive Explanations method. Experiments were performed on multiple public datasets. The results demonstrate that the proposed model achieves an accuracy of 96.8% , a precision of 96.2% , a recall of 97.7% , an F1-score of 96.9% , and an AUC value of 0.989. Its performance is significantly superior to that of other machine learning models. |
| Keywords: machine learning improved whale optimization algorithm ensemble technique heart disease prediction interpretability |
