Attribute importance measurement based on distributed identifiable matrices
A new calculation method is proposed for measuring the importance of attributes in target information sys-tems.Firstly,the conditional attribute set is reduced based on the identifiable attribute matrix.Then,according to the defini-tion of the lower approximation set in the variable precision rough set model,the formula for measuring the importance of attributes is given.Finally,for the given threshold β,the importance of each attribute in the reduced attribute set is calculat-ed separately.Example analysis shows the effectiveness and feasibility of this method.
Distributed identifiable matrixRough set modelDistribution reductionAttribute importance
万方数据
维普
