摘要
决策树因其形状像树且又能用于决策故被称为决策树,是通过机器学习,从一系列无秩序、无规则的逻辑关系中推理出一套分层规则,将结局按照概率分布的树形图表达,从而进行精确预测或正确分类。现系统综述了决策树在中医药领域的应用现状,发现决策树在疾病风险评估、中医病证的诊断、辨证分型、中药药性或不良反应的预测、证候与理化指标的关联、预后评估和成本-效果分析等方面均有所应用,且其分类和预测结果较为准确,值得今后进一步研究并推广应用。
A decision tree is called this name because it is shaped like a tree and it can be used for decision-making.It is a set of hierarchical rules inferred from a series of disordered and irregular logical relations by machine learning,and the outcomes are expressed according to the tree graph of probability distribution,so as to accurately predict or correctly classify.This paper systematically summarizes the present situation of the application of decision tree in the field of traditional Chinese medicine,and found that the decision tree in used in disease risk assessment,diagnosis of TCM diseases and syndromes,syndrome differentiation of TCM,the prediction of Chinese medicinal properties or adverse reactions,syndrome differentiation associated with physical and chemical indicators,prognostic evaluation and cost effect analysis and so on.The classification and prediction results by decision tree are more accurate,and is worth for further research and application in the future.
作者
马红丽
徐长英
杨新鸣
MA Hongli;XU Changying;YANG Xinming(The First Affiliated Hospital of Heilongjiang University of Chinese Medicine,Harbin 150040,China;Heilongjiang University of Chinese Medicine,Harbin 150040,China)
出处
《世界中医药》
CAS
2021年第17期2648-2651,2656,共5页
World Chinese Medicine
基金
国家自然科学基金青年基金项目(82004403)
2018年度黑龙江省普通高等学校青年创新人才项目(UNPYSCT-2018223)
黑龙江省自然科学基金面上项目(JJ2018ZR0618)。
关键词
决策树
数据挖掘
机器学习
中医药
预测模型
树形图
分类
风险评估
Decision tree
Data mining
Machine learning
Traditional Chinese medicine
Prediction model
Tree diagram
Classification
Risk assessment
