The Based on Rough Set Theory Development of Decision Tree after Redundant Dimensional Reduction

Abstract

Decision tree technologists have been examined to be a helpful way to find out the human decision making within a host. Decision tree performs variable screening or feature selection. It requires relatively lesser effort from the users for the preparation of the data. In the proposed algorithm firstly we have undertaken to minimize the unnecessary redundancy in the decision tree, reducing the volume of the data set decision tree is a fabrication through rough set. The main advantage of rough set theory is to press out the vagueness in terms of the boundary region of a set. Rough sets do not need the primitive conditions to decide the boundaries on time. The algorithm reduces a complexity and improve accuracy, then increase. The result experiment of better accuracy and diminished tree of the complexity proposed in this algorithm.