摘要
决策树作为从大规模数据中探索概念构成的代表,是弱化模型结构仅从数据出发构建概念的典型,所以决策树作为数据挖掘的典型技术得到了广泛的应用。根据临床检验资料信息,利用决策树模型建立2型糖尿病预测模型,为能更准确地诊断2型糖尿病提出理论依据。首先,搜集数据并进行预处理;然后,利用R语言编程构造ID3算法和CART算法的分类模型;再通过对ID3算法和分类回归树(CART)算法研究相比较,分析对比每个单一算法的性能和挖掘收集到的糖尿病数据;最后,比较2种方法的准确率,对预测结果进行评估。构建的诊断模型都具有较高的预测准确度,且CART模型优于ID3模型,对预测2型糖尿病的患病风险具有一定的临床参考价值。
According to the clinical data,aprediction model of type 2 diabetes was established by using decision-making tree model,which provides a theoretical foundation for more accurate diagnosis of type 2 diabetes.As a representative of exploring concept structure from big data,it is a typical example of weakening model structure only using data to construct concept.Therefore,as a typical technology of data mining,decision-making tree has been widely applied.So in this paper,it is used as the research method to establish the model.First collect and carry out preprocessing data,Then,R is used to construct the classification model of ID3 and CART.And we compare the ID3 and the classified regression tree（CART）,analyze and compare the performance of each single algorithm and mining the diabetes data collected.Finally,the accuracy of the two methods is compared and the prediction results are evaluated.The diagnostic models constructed in this paper have high prediction accuracy.And the CART model is superior to the ID3 model.It has a certain clinical reference value for predicting the risk of type 2 diabetes.
作者
杨光
马尔丽
YANG Guang;MA Erli(College of Mathematics and Systems Science,Shenyang Normal University,Shenyang 110034,China)
出处
《沈阳师范大学学报（自然科学版）》
CAS
2018年第3期233-236,共4页
Journal of Shenyang Normal University:Natural Science Edition
基金
辽宁省教育厅科学研究一般项目(LQN201710)
