Research on an Attribute Section Method for Decision Tree
Guo Kai · Journal of Taiyuan University of Technology · 2011
As ID3 algorithm is inclined to choose the attributes with more values and less effect on classification as the test attributes,the OneR algorithm is introduced to select the relevant attribute subsets to classify the dataset,reduce the effect of irrelevant and repeated attributes.The experimental results show that the improved solution has higher classification accuracy and less classification time over the ID3 algorithm,overcomes the problem of value deviation,lastly optimizes the classification results.