A high-dimensional attribute reduction method modeling and evaluation based on green economy data: evidence from 15 sub-provincial cities in China

Data play an increasingly crucial role in decision evaluation. However, the noise and redundant information in the data create confusion to the decision makers. To solve this problem, this paper creates a new attribute reduction model based on technique for order preference by similarity to an ideal solution (TOPSIS), grey correlation analysis and coefficient of variation approaches. First, we obtain the time weights of panel data in different years by TOPSIS and then transfer the panel data into a cross-sectional data matrix. Second, we delete the overlapping attributes by grey correlation analysis method. Third, we use the coefficient of variation to select the attributes with the highest information content. Finally, the proposed attribute reduction model has been varied based on the green economy evaluation data of 15 sub-provincial cities in China. The experimental findings provide decision-making reference for the local government policymakers to adjust or formulate green economic development strategies.