Heuristic Algorithm Based on Standard Deviation Attribute Significance in Supervised Neighborhood Rough Set
LIU Xudong
Abstract：As a significant model in rough set theory,the supervised neighborhood rough set has been paid much attention be-cause of its better discriminating performance.However,when deriving reduct in supervised neighborhood rough set,most research-es will cause massive time consumption because of repeatedly computing the significance of candidate attributes.To fill such a gap,a heuristic algorithm based on the standard deviation attribute significance is proposed.The attribute significance is sorted according to the degree of the dispersion of the samples,so as to reduce the traversal scale of attributes in the process of deriving reduct.Exper-imental results over 12 UCI data sets show that the proposed algorithm can effectively decrease the elapsed time of deriving reduct and improve the classification performance simultaneously over the other three algorithms.
Keywords：acceleration strategyattribute reductionattribute significancestandard deviationsupervised neighborhood rough set
Publication Date：2025-01-19
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：6( 15-20 )