Information processing method based on ANP-DEA-BP neural network model

Abstract

The invention provides an information processing method based on an ANP-DEA-BP neural network model, and belongs to the technical field of information processing, a DEA model and an ANP method are integrated, and the method can be used for evaluating the efficiency of various input and output decision units, so that the efficiency of decision units with complex production relations is evaluated. And the resource use condition of the decision-making unit can be further known. After the ANP-DEA model is used for obtaining the comprehensive value, the value is input into the BP neural network, and the influence of subjective factors such as subjectivity, uncertainty and fuzziness of people during evaluation is reduced, so that the accuracy and credibility of an evaluation result are improved.

Description

technical field [0001] The invention belongs to the technical field of information processing, and in particular relates to an information processing method based on an ANP-DEA-BP neural network model. Background technique [0002] At present, the field of information processing technology can solve various complex system problems for decision makers, but it also has some shortcomings. For example, the interaction between different decision-making levels or the same level is not considered; explicit expressions between input and output indicators need to be set in advance, and there are many subjective factors, which make the conclusions highly subjective. [0003] The multi-layer forward BP neural network evaluation model is the most widely used neural network form of the multi-index evaluation model. It has the general advantages of neural networks, but it is not perfect. From a mathematical point of view, the traditional BP neural network is an optimization method that s...

AI Technical Summary

Problems solved by technology

From a mathematical point of view, the traditional BP neural network is an optimization method that solves a complex nonlinear problem. The weights of the network are gradually adjusted along the direction of local improvement, which will cause the algorithm to fall into local The extreme value, the weight converges to the local minimum point, which leads to the failure of network training; because the BP neural network algorithm is essentially a gradient descent method, the objective function it needs to optimize is very complicated, so the "sawtooth phenomenon" will inevitably appear. ", which makes the BP algorithm inefficient; and because the optimized objective function is very complex, it will inevitably appear some flat areas when the neuron output is close to 0 or 1. In these areas, the weight error changes little, Brings the training process to a near standstill

Images