Knowledge granularity incremental attribute reduction algorithm based on neighborhood relation

In order to study incremental attribute reduction in neighborhood information system, an incremental attribute reduction algorithm of neighborhood information system is constructed by using neighborhood knowledge granularity. Firstly, the knowledge granularity of information system is extended in the neighborhood information system, and the neighborhood knowledge granularity is proposed. Secondly, the incremental updating mechanism of neighborhood knowledge granularity is studied for the case of attribute increase. Finally, the corresponding incremental attribute reduction algorithm is designed based on this mechanism. Experimental results show that the incremental algorithm has high dynamic attribute reduction performance.