A New Fuzzy Neural Network Algorithm for Rural Public Service Performance Evaluation

The defects of BP neural network, such as low convergence speed, falling into local minimum easily, bad generalization ability, can depress the calculation accuracy of BP neural network and damage its practical effect. So the research of improving BP neural network has great theoretical and practical significance. The paper advances a new fuzzy neural network algorithm to overcome the defects of original BP neural network algorithm and evaluate rural public service performance. First, the paper designs a new calculation structure based on fuzzy and BP neural network theory, and selects new selftraining methods for the improved fuzzy neural network algorithm; Second, the performance of the advanced algorithm is also analyzed form four aspects in theory; Finally, based on analyzing and constructing the evaluation indicator system, the improved fuzzy neural network algorithm is applied to evaluate rural public service performance and the experimental results show that the superiorities of the improved algorithm include high evaluation accuracy, fast convergence speed, small oscillation, simple algorithm process.