Study on grey fixed weight clustering decision-making based on interval numbers

This paper is an emphasis on some problems of grey fixed weight clustering decision-making based on interval numbers and gives definitions of ideal attribute interval numbers with benificial indicators and cost indicators respectively. At the same time, ideal attribute deviation degree is defined and the matrix of ideal attribute deviation degree is converted from the matrix of interval numbers by the definition. Then weights of clustering indicators are calculated according to the concept and characteristic of information entropy. This method which is presented in the paper could deal with the conditions better when meanings and dimensions of indicators are different and numerical values have big disparity each other and confirm weight of clustering indictors with interval number. Finally, an example is presented to illuminate the feasibility and rationality of the method.