A New Method for Classification of Datasets for Data Mining.

摘要: Decision tree is an important method for both induction research and data mining, which mainly used model classification prediction. ID3 algorithm the most widely in decision so far. In this paper, shortcoming of ID3's inclining to choose attributes with many values discussed, then a new improved version ID3. our proposed are divided into groups we apply selection measure 5 these groups. If information gain not good again divide These steps done until get classification/misclassification ratio. The algorithms classify sets more accurately efficiently.