One New Algorithm for Intuitiontistic Fuzzy-rough Attribute Reduction
Huicheng Yang · Journal of Chinese Computer Systems · 2011
This paper proposed one new method of attribute reduction integrating the information entropy with intuitionistic fuzzy rough sets.The information entropy was generalized in intuitionistic fuzzy rough sets so that it could be used to value importance and dependency of attribute under intuitionistic fuzzy circumstance.The algorithm we proposed based on the criteria of the importance and dependence of attribute by the addible and erasable stepwise-regression.With the unvaried classification accuracy,one simpler decision attribute is obtained.At last,simple examples illustrated that decision rule based on attribute reduction which is obtained by the algorithm we proposed is simpler and more perfect set in multi-attribute reduction.