Stable Attribute Cluster Reduction Mechod Based on Density Peaks Clustering
ZHAO Dasheng
Abstract：Attribute reduction is one of the most important research directions in the field of the rough set.As an attribute re-duction method based on the heuristic algorithm,the attribute group method has attracted much attention because it can reduce the time consumption of deriving reducts.However,due to the randomness and instability of K-means clustering in attribute group method,the generated attribute groups are unstable,which further affects the stability of reducts.Therefore,to solve this problem,an attribute reduction method with a stable attribute group based on density peaks clustering is proposed.Specifically,in the pro-cess of searching reducts,candidate attributes are scanned based on stable attribute groups generated by density peaks clustering,so as to reduce the number of traversal times of attributes and improve the stability of reduct.Experimental results on 10 UCI data sets show that the proposed algorithm can not only generate more stable reducts than the traditional attribute group method,but also improve the efficiency of searching reducts effectively without reducing the classification performance.
Keywords：attribute groupattribute reductiondensity peaks clusteringneighborhood rough setstability
Publication Date：2025-06-20
Online Publishing Date：2025-09-23(First online date of this platform, not the publication date of the document)
Pages：6( 1527-1532 )