A New Algorithm for Attribute Reduction Based on Discernible Matrix
Sichun Yang
Abstract
Sichun Yang
Abstract
Attribution reduction is one of the important topics in the research on rough set theory.But the attribution reduction is a NP problem,and it needs to be realized by knowledge of elicitation method.Proposes the method that can calculate the positive region of different condition attribution combination relatively decision attribution.A new approach to compute the core attribute is also put forward in this paper.On the basic of this,the new algorithm for attribute reduction based on discernible matrix is proposed.The algorithm can get the smallest attributes quickly and be realized easily.It is realized that attribute reduction is step with rules collection.And it is proved to be workable in the practice.
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Attribution reduction is one of the important topics in the research on rough set theory.But the attribution reduction is a NP problem,and it needs to be realized by knowledge of elicitation method.Proposes the method that can calculate the positive region of different condition attribution combination relatively decision attribution.A new approach to compute the core attribute is also put forward in this paper.On the basic of this,the new algorithm for attribute reduction based on discernible matrix is proposed.The algorithm can get the smallest attributes quickly and be realized easily.It is realized that attribute reduction is step with rules collection.And it is proved to be workable in the practice.
Key concepts: Rough set, Reduction (mathematics), Computer science, Attribution, Algorithm, Matrix (chemical analysis), Core (optical fiber), Set (abstract data type)