A dynamic rule-based classification model via granular computing

As an effective tool for data representation and processing, granular computing has been incorporated into formal decision contexts finding reducts to achieve the task of mining rules. However, classification performance rules not evaluated, this type method is suitable dynamic data. To solve problem, current study updates evaluates obtained in terms performance. Concretely, we first give a theoretical analysis updating then present novel rule-based model (DRCM) based on mechanism. Finally, discuss feasibility proposed compare it with several popular algorithms. The conducted experiments demonstrate that can improve ability certain extent DRCM better some consistent datasets.