基于粒度熵的知识约简算法及其应用

Granular computing theory is a relatively new soft computing tool to analyze and deal with incomplete information effectively. In this paper, we introduce the concept of knowledge granulation, granularity entropy, relatively granularity entropy, the importance of attribute and their computational methods. Also, we propose the knowledge reduction algorithm based on granularity entropy and use it in fault diagnosis of power transformer. Finally, the validity of proposed algorithm is proved by the application of practical database.