Rough set-based decision tree construction algorithm

Abstract

We apply rough set theory to obtain knowledge from the construction of a decision tree. Decision trees are widely used in machine learning. A variety of methods for making decision trees have been developed. Our algorithm, which compares the core attributes of objects and builds decision trees based on those attributes, represents a new type of tree construction. Experiments show that the new algorithm can help to extract more meaningful and accurate rules than other algorithms. © Springer-Verlag Berlin Heidelberg 2007 Rough set, Decision Tee, Core.