An ensemble classifier based on attribute measurement of rough sets

From the viewpoint of the attribute measurement of rough sets,a new attribute measurement based on the hybrid metric mechanism was provided to accurately evaluate the significance of attributes. This proposed attribute measurement analyzes the significance of attributes from different levels of information granularity. In addition,a parameter weighting factor was introduced to the attribute measurement according to the characteristics of data distribution. On this basis,an ensemble classifier was constructed based on the proposed attribute measurement mechanism in rough sets. Experimental results and comparative analysis show that the proposed method can effectively reduce the attribute dimension of data. Compared with the single attribute measurement,the proposed method has a better classification performance.