Analysis of Data Mining Classification with Decision Tree Technique
The diversity and applicability of data mining are increasing day to day so need to extract hidden patterns from massive data. The paper states the problem of attribute bias. Decision tree technique based on information of attribute is biased toward multi value attributes which have more but insignificant information content. Attributes that have additional values can be less important for various applications of decision tree. Problem affects the accuracy of ID3 Classifier and generate unclassified region. The performance of ID3 classification and cascaded model of RBF network for ID3 classification is presented here. The performance of hybrid technique ID3 with CRBF for classification is proposed. As shown through the experimental results ID3 classifier with CRBF accuracy is higher than ID3 classifier.
