基于知识粗糙度的决策树生成方法

决策树是一种重要的数据分类方法,测试属性的选择直接影响到决策树中结点的个数和深度,本文提出了一种基于知识粗糙度的方法.通过比较我们发现:在决策树的构造上,粗集理论中知识粗糙度的方法计算量较小,构造的决策树比经典ID3算法简洁,并且具有较高的分类精度.

A Knowledge Roughness Based Approach to Decision Tree Construction

Decision Tree is one of the most important method for datamining,the choice of testing attribute will touch the depth and nodes of decision tree.In this paper we bring forward a new approach to decision tree construction based on knowledge roughness.By comparation we find that the new method has less compute capacity and constructed decision tree is concision with higher classify precision.