Open AccessModel of Normal University Students’ Education Quality Based on Artificial Neural Networks
Author(s) -
Shaofeng Wang,
Yue He,
Haitao Lin
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1828/1/012146
Subject(s) - artificial neural network , computer science , artificial intelligence , machine learning , python (programming language) , sample (material) , valuation (finance) , matlab , quality (philosophy) , reliability (semiconductor) , data mining , philosophy , chemistry , power (physics) , physics , finance , chromatography , epistemology , quantum mechanics , economics , operating system
As a main form of checking out learning and teaching effects, exam has traditionally been used in schools. Accurate and objective assessment of students’ learning effects through examination performance is an important part of evaluation of students’ education. This paper proposes a comprehensive evaluation model of evaluating examination performance based on quality of examination paper, explores the application of artificial neural network technology in students’ comprehensive education evaluation, focusing on BP neural network, some works can also be introduced by Python. The process is to compose input vector of BP neural network using the four evaluating indicators of quality of examination paper, namely reliability, validity, difficulty and discrimination, and raw score, namely the students’ score of examination paper. Take valuation namely the quantitative value of learning effect as the output vector of BP neural network. Design a reasonable network structure and training sample, put the training sample in the network for processing till until systematic error meets the specified requirements. By so, the obtained network model is the required comprehensive evaluation model of examination performance. At last, the paper analyzes the simulation of the evaluating model’s feasibility using MATLAB software, obtains satisfying results, and proves the feasibility of the proposed network model.