Ecological civilization evaluation model based on neural network

With the determination of the dual-carbon target, the evaluation of the degree of ecological civilization construction can not simply rely on the weighted calculation of specific elements. At present, expert assessment is the main method to evaluate ecological civilization degree. This paper puts forward an ecological civilization model which is composed of three aspects: "ecological environment", "living standard" and "social harmony". From the perspective of convenient implementation evaluation, the author selects the easily available "Engel coefficient", "GDP", "disposable income of residents", "forest coverage" and "carbon sequestration" to evaluate the degree of ecological civilization construction. The neural network is used to simulate the expert evaluation process to realize the complex nonlinear mapping relationship in the actual evaluation. On the basis of determining the input and output values of the neural network, the neural network structure were analyzed theoretically. The practical calculation of the ecological civilization construction degree of 5 counties in northwest China shows that the neural network proposed in this paper can simulate the expert judgment process well. The experiment of confusion matrix shows that the neural network proposed in this paper agrees with the expert judgment up to 80%. Considering that the degree of ecological civilization construction is subjective to a certain extent and the opinions of different experts may be different, the neural network actually has a certain ability to evaluate the ecological civilization.