基于心脏病患病情况的评估模型研究.docx

目录 TOC \o 1-3 \h \u 1 引言 1 1.1 研究背景及意义 1 1.2 国内外研究现状 1 1.2.1 国外研究现状 1 1.2.2 国内研究现状 2 1.3 课题发展趋势 3 1.4 研究思路 3 1.5 研究方法 4 2 数据描述 5 2.1 数据集介绍 5 2.2 分类特征值比 5 2.3 分类特征值与目标值之比 9 2.4 连续特征值及患病情况的分布 10 3 模型建立 12 3.1 Logistic回归 12 3.2 随机森林 12 3.3 K近邻分类 13 3.4 朴素贝叶斯 13 4 模型评估与分析 14 4.1 数据集预处理 14 4.2 评估指标 14 4.3 模型评估 17 4.4 结果分析与对比 21 5 总结与展望 23 5.1 总结 23 5.2 不足与展望 23 参考文献： 25 致谢 26 基于心脏病患病情况的评估模型研究 摘要：当今社会，科学技术的高速发展给人们带来了许多便利，人们的生活水平得到提高和改善，然而不规律的作息时间正是许多人当下的生活写照，其弊端也逐渐显露。最为突出的就是心脏病，是死亡率较高的一种疾病，不仅在我国，甚至是世界范围内都对心脏病有着重要的研究。但是医疗研究是一个繁琐而又复杂的过程，对准确率的要求极为严格，在这种情况下，可以通过使用机器学习完成这个复杂的过程，利用它来研究预测和分析数据。本文对心脏病患病情况作出研究，通过建立单一的评估模型——Logistic回归模型、K近邻分类、朴素贝叶斯、随机森林，根据模型的评估指标的对比，对模型进行评估分析，结果表明Logistic回归的效果最佳。 关键词：心脏病；机器学习；评估模型；评估指标 A model study based on the assessment of heart disease Abstract:In todays society,the rapid growth of science and technology has brought a lot of convenience to people, and their living standards have been improved and improved. However, the irregular schedules are just a portrayal of many peoples lives nowadays, and their disadvantages are gradually revealed.The most prominent is heart disease, a disease with a high death rate. Not only in our country, but even in the world, there are important studies on heart disease. However, medical research is a complicated and complicated process with extremely strict requirements on accuracy. In this case, this complex process can be completed by using machine learning to study prediction and analyze data. In this paper, the incidence of heart disease was studied. By establishing a single evaluation model -- Logistic regression model, k-nearest neighbor classification, naive bayes, and random forest, by comparing the evaluation indicators of the model, the evaluation and analysis of the model shows that the effect of Logistic regression is the best. Key words: Heart disease; Machine learning; Evaluation model; Evaluation indicators PAGE