论文部分内容阅读
决策树，作为一种强有力的数据分析工具，正在引起统计学家极大的兴趣。本文介绍一种新的决策树构建的方法及其步骤。该法以样品中信息量为指导，以训练样本为基础，通过选择性试验，即选择好的属性作为树根、子树根、不断重复，最终建立决策树。同时，为了帮助读者更好地理解整个建树过程，文章以人体体型分类为例，对决策树在分类方面的应用进行了初步探讨，并且给出了人体体型分类的决策树图和决策树构建过程示意图。
Decision trees, as a powerful data analysis tool, are causing great interest from statisticians. This article describes a new decision tree construction method and its steps. This method is based on the amount of information in the sample, based on the training sample, and through selective tests, selecting good attributes as roots, sub-tree roots, and repeating, finally establishing a decision tree. At the same time, in order to help readers better understand the entire construction process, the paper takes the human body classification as an example to conduct a preliminary discussion on the application of decision trees in classification, and gives a decision tree and decision tree construction process for human body shape classification. Schematic diagram.