Research on Regional Basic Education Quality Assessment Based on Deep Convolutional Neural Network

Abstract

At present, the informatization of basic education quality assessment has become a hot topic in the field of education and is playing an increasingly important role. Based on the theory of deep convolutional neural network, this paper adopts the methods of mathematical analysis and experimental research to construct a regional basic education quality assessment model. The model solves the data informatization problem of education quality assessment. In the simulation process, two key modules of data self-assessment and expert assessment of the deep convolutional neural network are realized by ASENET+SQL SERVER, and the assessment results are integrated by using the weighted average method and the fuzzy comprehensive assessment method. The experimental results show that the quantitative analysis of the quality assessment is carried out by using the logic and support relationship, and the results of comprehensive qualitative analysis and quantitative analysis are realized and segmented when the threshold level is 9, the MIOU obtains the highest value of 0.7501, and the MIOU of the multi-stage method of the quality evaluation model proposed in this paper is 0.8116, which is 6.15% points higher than the traditional multi-stage algorithm, which effectively improves the current stage area quality of basic education.