Classification and Pruning Strategy of Knowledge Data Decision Tree Based on Rough Set

Abstract

Decision tree classification is a method to deduce classification rules from irregular and disordered training sample sets. In this method, the top-down comparison method is used to different attribute values. The basic principle of the reduction algorithm in rough set is to find out the minimum set of related attributes with the same decision or resolution capability of the original data in the generalization relation by seeking the importance of the attributes and to sort them, so as to realize the information reduction. The paper presents classification and pruning strategy of knowledge data decision tree based on rough set.