Quantitative Evaluation of NDE Reliability Based on Back Propagation Neural Network and Fuzzy Comprehensive Evaluation

There are some problems in the quality evaluation of network distance education, such as the large error in the quantitative results of the evaluation indicators, the low accuracy of the quality evaluation and the large error in the indicator's weight distribution. Therefore, a quantitative evaluation method of NDE reliability based on Back Propagation (BP) neural networks and fuzzy comprehensive evaluation is proposed. From the perspective of students, teachers and schools, the primary and secondary indicators of teaching evaluation are constructed, and the reliability evaluation indicators system of teaching quality is established. The fuzzy comprehensive evaluation method is used to calculate the membership degree of the evaluation indicators, the consistency verification method is used to determine the consistency of the indicator's attributes, and the quantitative research of the indicators data is completed by the weight calculation method to improve the accuracy of the quantitative results. BP neural network is used to construct the input layer, hidden layer and output layer of network distance education quality reliability evaluation indicators, calculate the weight of evaluation indicators, transform the input evaluation indicators data through linear conversion function, construct the indicators evaluation model, and correct the evaluation results with the help of loss function to improve the accuracy of quality evaluation results. The experimental results show that the evaluation result error of the proposed method is relatively small, with an error of less than 0.3%, and the highest evaluation accuracy reaches 96.3%, which can effectively improve the quantitative evaluation effect of the reliability of network remote teaching quality.