摘要
剪枝过程是决策树分类学习中的重要环节,能够简化决策树并提高决策树的泛化能力,避免对训练数据集的过适应。在PEP算法的基础上,本文提出了一种改进的决策树剪枝算法IPEP,实验结果表明,该算法剪枝效果较PEP算法更好。
Pruning is an important part of decision tree induction, which can simplify and populate decision trees and avoid the over-fitting question. In this paper, we propose an improve algorithm for decision tree pruning based on PEP, named IPEP. The result of experiments shows that IPEP has a better pruning effect than PEP.
出处
《科技广场》
2010年第6期49-51,共3页
Science Mosaic
