Font Size: a A A

Research And Application Of School Education Quality Evaluation Method Based On Compulsory Education Balanced Indicators

Posted on:2020-05-09Degree:MasterType:Thesis
Country:ChinaCandidate:X Q ChenFull Text:PDF
GTID:2417330596479113Subject:Software engineering
Abstract/Summary:PDF Full Text Request
Education is not only a judging criterion for the country's soft power,but also a platform for the concentrated display of national culture.In order to promote the balanced development of compulsory education,realize compulsory educational equity and strengthen the educational concept of "high grade,fine connotation,strong characteristics",the state continues to vigorously promote education reform and development,among them,the school education quality assessment is particularly important.At present,most of the current research on the school education quality evaluation is about qualitative analysis,and there are few quantitative evaluation studies.Moreover,the current evaluation system has some problems.For example,evaluation indicators are not incomplete and evaluation methods are imperfect.In order to reflect the school education quality comprehensively and effectively,the research focus on the selection of evaluation indicators and optimization of the evaluation methods,and realizes the design and development of the application system of school education quality assessment model,the main research contents are as follows:1.Principal component analysis(PCA)method is used to explore the selection of evaluation indicators.Supplementing the compulsory education balanced indicator,the PCA method extracts the principal components from the new evaluation indicators and old evaluation indicators,and calculates the evaluation scores of the integrated principal components,thus obtaining the evaluation results of the school education quality.The experimental results show that the new evaluation indicators can reflect the quality of school education more than the old evaluation indicators,thus reflecting the scientific and feasibility of the new evaluation indicators.2.The BP neural network evaluation model is used to achieve the school education quality assessment.The evaluation model has a good nonlinear mapping approximation ability,which effectively avoids the principal component analysis method linearly combines the evaluation indicators simply.Network training and learning are performed on the evenly distributed and normalized experimental data to determine the network structure,activation function,loss function and type of optimizers of the BP neural network.BP neural network over-fitting and under-fitting phenomena are analyzed and the corresponding solutions are proposed.The experimental results show that the BP neural network evaluation model can obtain better evaluation results,and can solve the problem of ambiguity and randomness of weight determination.3.A cross-validation GA-BP neural network assessment model is proposed to evaluate the school education quality.In order to overcome the problem of the BP neural network evaluation model,such as slow network learning speed,easily falling into local minimum value and unstable network performance,the genetic algorithm is applied to optimize the initial parameters of BP neural network.For small amount of experimental data,the GA-BP neural network is proposed to realize cross-validate with experimental data set,and experimental results of cross-validation are used as an important reference for school education quality assessment.The global optimization of the genetic algorithm and the local optimization of BP neural network ensure that BP neural network converges to global optimality,and cross-validation solves the problem of small experimental data set.The two factors together improve the accuracy of school education quality assessment.4.The proposed model and method are applied to a city education decision support system.The software engineering ideas and methods are used to establish overall architecture,interaction diagram and E-R diagram of the system.The test implementation and overall effect diagram of the system are given.The evaluation method is proposed in this research that has been successfully applied to the evaluation and decision support system of compulsory education in a certain city,has been played an important reference role in the teaching evaluation and scientific decision-making of the city and county education administrative department.It has been highly recognized by users.
Keywords/Search Tags:School education quality evaluation, Principal component analysis, BP neural network, Cross-validation, GA-BP neural network
PDF Full Text Request
Related items