Scholarlink

<p>Passing college English test Band 4/6 (CET-4/6) is closely related to the future development of college students, and predicting the pass status can help students and colleges more clearly grasp the situation of English learning. This paper proposes a combination of genetic algorithm (GA) and particle swarm optimization (PSO) for the optimization of the back-propagation neural network (BPNN) to obtain the initial weight threshold. Some non-English major undergraduate students enrolled in 2022 and 2023 from Zhengzhou Shengda University were selected as samples to predict their pass status (pass/not pass) in CRT-4/6. The results showed that methods such as logistic regression and support vector machine (SVM) performed poorly in predicting students’ pass status in CET-4/6. The GA-PSO-BPNN method achieved an accuracy of 0.7864 and an F1 score of 0.7815 for predicting the CET-4 pass status, and it achieved an accuracy of 0.8112 and an F1 score of 0.8157 for predicting the CET-6 pass status, which was superior to the other algorithms. The results demonstrate the reliability of the GA-PSO-BPNN method in predicting the CET-4/6 pass status of college students, which can be applied in practice.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Predicting Chinese college students’ pass status in CET-4 and CET-6 using machine learning

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月27日

查看全文
  • Jinna Zhang

摘要

Passing college English test Band 4/6 (CET-4/6) is closely related to the future development of college students, and predicting the pass status can help students and colleges more clearly grasp the situation of English learning. This paper proposes a combination of genetic algorithm (GA) and particle swarm optimization (PSO) for the optimization of the back-propagation neural network (BPNN) to obtain the initial weight threshold. Some non-English major undergraduate students enrolled in 2022 and 2023 from Zhengzhou Shengda University were selected as samples to predict their pass status (pass/not pass) in CRT-4/6. The results showed that methods such as logistic regression and support vector machine (SVM) performed poorly in predicting students’ pass status in CET-4/6. The GA-PSO-BPNN method achieved an accuracy of 0.7864 and an F1 score of 0.7815 for predicting the CET-4 pass status, and it achieved an accuracy of 0.8112 and an F1 score of 0.8157 for predicting the CET-6 pass status, which was superior to the other algorithms. The results demonstrate the reliability of the GA-PSO-BPNN method in predicting the CET-4/6 pass status of college students, which can be applied in practice.