The choice of attributeselection metric to split has an important impact on the shape and the depth of the resulting decision tree.
在根据入侵规则构造决策树时，所依据的分类属性选择标准对决策树的形状和深度有很大的影响。
2
The abstract channel model for learning from examples is presented and a new attributeselection measure (channel capacity) is introduced.
本文提出了示例学习的抽象信道模型，引入一个新的特征选择量——信道容量。
3
We adopt a way of attributeselection based on word entropy, use vectors which are represented by word frequency, and deduce its corresponding Bayesian formula.