Multivariate Decision Tree Algorithm Based on Rough Set

The decision tree is a widely used classification model and inductive learning method based on examples. It is characterized by the simple classification rules, easy understanding for users and so on, but we also can see some disadvantages in certain situations. The paper puts forward the multivariable decision tree algorithm which based on a rough set to a combination of rough sets theory and decision tree algorithm. The multivariable decision tree algorithm has reduced the complexity of decision tree while not affect the readability of the classification rules. Experimental analysis has witnessed the feasibility and efficiency of the algorithm.