| 决策树后剪枝算法的研究 |
| |
| 引用本文： | 范洁,杨岳湘.决策树后剪枝算法的研究[J].湖南广播电视大学学报,2005(1):54-56. |
| |
| 作者姓名： | 范洁 杨岳湘 |
| |
| 作者单位： | 1. 湖南广播电视大学
2. 国防科技大学计算机学院 |
| |
| 摘 要： | 决策树学习策略广泛应用于模式识别和机器学习等领域，用来解决与分类相关的问题。决策树剪枝的作用是简化决策树，提高决策树的泛化能力，避免对训练集的过适应，是决策树学习中的重要研究内容。本详细描述了常用的四种后剪枝算法，分析了后剪枝技术的研究与发展现状，为具体应用中选择剪枝算法提供了一定的理论基础。
|
| 关 键 词： | 决策树 后剪枝 过适应 过剪枝 欠剪枝 |
| 文章编号： | 1009-5251(2005)01-054-03 |
Researches on Methods for Post-pruning Decision Trees |
| |
| FAN Jie,YANG Yue-Xiang.Researches on Methods for Post-pruning Decision Trees[J].Joournal of Hunan Radio and Televistion University,2005(1):54-56. |
| |
| Authors: | FAN Jie YANG Yue-Xiang |
| |
| Institution: | FAN Jie YANG Yue-Xiang |
| |
| Abstract: | Decision tree learning strategy have long been popular in pattern recognition,machine learning,and other disciplines for solving problems concerned with the classification.Pruning is an important domain of decision tree induction,which can simplify and populate decision trees and avoid overfitting.In this paper,we describe four well-known post-pruning methods with the aim to understanding their theoretical foundations and supplying references in practice. |
| |
| Keywords: | Decision tree Post-pruning methods Overfitting Overpruning Underpruning |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
