基于LightGBM的心血管疾病预测模型研究

Research on Cardiovascular Disease Prediction Model Based on LightGBM

应用机器学习对心血管疾病预测尤为重要。本文构建一种基于LightGBM的心血管疾病预测模型,并利用Kaggle平台的疾病数据集进行训练。该模型通过精确率、召回率和F1值3个指标进行性能评估,并将其与Logistic回归、梯度提升决策树(Gradient Boosting Decision Tree,GBDT)、随机森林进行对比实验。实验结果表明,本文构建的模型优于其他算法。利用KNN对该模型进行优化,最后通过10折交叉验证分析效果,结果表明经过K最邻近算法(K-NearestNeighbor,KNN)优化后,提高了该模型的预测能力。

The application of machine learning is particularly important for cardiovascular disease prediction.In this paper,a LightGBM-based cardiovascular disease prediction model is constructed and trained using a disease dataset from the Kaggle platform.The model’s performance is evaluated by three metrics:accuracy,recall and value,and it is compared with Logistic Regression,Gradient Boosting Decision Tree(GBDT),and Random Forest for experiments.The experimental results show that the model constructed in this paper outperforms the other algorithms.The model is optimized using KNN,and finally the analysis effect is verified by 10-fold crossover,and the results show that the pr ediction ability of the model is improved after K-NearestNeighbor(KNN)algorithm optimization.