Research And Application Of Quality Evaluation Algorithm For Higher Engineering Education Based On Quasi-neural-network Architecture

Higher engineering education is the main channel for the training of engineering qualified scientists and technicians.The engineering qualified scientists and technicians directly determines the industrial competitiveness and national economic level of a country.In 2016,China officially joined the international engineering education "Washington Accord" Organization,marking that the quality certification system of engineering education has achieved international substantive equivalence,and that the quality standards of engineering majors have reached international recognition,which is a major breakthrough in China's higher education.However,only by actively implementing the certification management mechanism and control measures,can we improve the quality of engineering education on the basis of the completion of equivalent requirements effectively.Among them,the construction of quality evaluation mechanism and feedback mechanism for the talent training system in accordance with the basic law of higher education is the fundamental support to ensure the quality improvement of higher engineering education.This paper takes the student-centered,outcomes-based education,continuous improvement and other core concepts in engineering education professional certification as the guide,and combines the basic laws of higher engineering education personnel training to build a quasi-neural network based higher engineering education quality evaluation models and algorithms.The application system is developed to achieve educational informatization,which provides effective tools for improving the quality of higher engineering education.The main research content has the following parts.1.According to the basic laws of the higher engineering education talent training system and the basic principles of neural network operation,the correspondence between the entity of the talent training system such as the curriculum system,training objectives,graduation requirements,and the input layer,hidden layer and output layer of the neural network is analysed.And the corresponding relationship between the inner relationship between the entities of the talent training system and the connection weight of the neural network nodes.Construct a quasi-neural network architecture to describe the logical relationship and operating mechanism between the multi-level subjects of the training system.The design of the architecture reflects the concept of OBE(Outcomes-based Education)in professional certification,and it provides a framework basis for designing a model and algorithm for higher engineering education quality evaluation of quasi-neural network.2.Propose a fuzzy evaluation algorithm for the quality of higher engineering education based on the architecture of quasi-neural network,which can effectively integrate the forward propagation mechanism,multi-level feedback mechanism and fuzzy quality evaluation mechanism in the talent training system.This algorithm provides a theoretical basis for rational analysis of the weak links in the cultivation of students' ability in the process of talent training.Among them,the forward propagation mechanism is based on process data,using the logic and supporting relationship between entities to quantitatively analyze entities.The fuzzy quality evaluation mechanism makes a comprehensive evaluation based on the results of forward propagation and multi-party feedback,and then carries out a qualitative evaluation of the entity.According to the talent training system,the multi-level feedback mechanism integrates the results of qualitative analysis and quantitative analysis,and it performs hierarchical feedback to provide a basis for the continuous improvement of the quasi-neural network architecture.3.Taking the higher engineering education quality evaluatin as the fundation of information construction,design the higher engineering quality evaluation system.Use the system to manage the entities and the relationships between the entities,and calculate the achievement degree of entities based on this.By comparing and evaluating the entity with the degree of analysis,it provides data support for the improvement of the talent training cultivation system.The system desgin is based on software engineering theory.At the same time,in order to achieve the separation of front and back ends and effectively improve the user experience,MVVM(Model-View-ViewModel)mode is used for design.In order to facilitate the management of the system at a later stage,during the system development process,standardized steps such as requirements analysis,conceptual design,detailed design,code implementation,and testing are performed in sequence.This paper uses a large amount of basic teaching data from the computer science and technology major of University of Jinan to conduct relevant experiments.The results show that the system developed based on the model and algorithm of quality evaluation,and quality evaluation model of higher engineering education can effectively realize the objective evaluation of the quality of talent cultivation,which is helpful to discover the weak links in teaching activities,and can provide an effective tool for improving the quality of higher engineering education in China at this stage.