Scholarlink

<p>Effective teaching plays a vital role in universities and colleges, creating an excellent success rate for students and institutions. An effective teaching process creates a positive learning environment that helps students understand the material and improves their academic achievement. During the learning process, frequent assessments are necessary to ensure high-quality education, and the teaching process is evaluated based on various factors. However, the learning process presents challenges related to standardized test scores. Hence, there is a need for more advanced methodologies that can account for multiple factors and provide more precise, valuable assessments of instructional efficacy. This study aims to address the issue by employing fuzzy logic and deep learning methods to assess the factors influencing teaching effectiveness in higher education institutions. The research uses data from multiple institutions, accounting for variables such as the student-to-teacher ratio, student academic achievement, and teacher expertise. The data gathered is subjected to deep learning techniques to analyze the intricate correlations between the input factors and teaching success. The fuzzy logic technique is applied during the analysis to address the inherent uncertainty and vagueness in the data effectively. The student’s performance and teaching experience are used to evaluate the impact of teaching. The model achieved 97.89% accuracy, indicating it outperforms other models, such as TLDL, SEM-ANN, and regular ANN methods, in predicting teaching performance. It additionally achieved an accuracy of 86.6%, a recall of 75%, an F1-score of above 80%, and a low error rate of 10.4%. These performance measures demonstrate that the FDL technique is robust and dependable in handling data uncertainty and extracting valuable insights from complex educational datasets. The successful results indicate that combining fuzzy logic with deep learning significantly enhances prediction performance, providing educational institutions with a valuable tool for evaluating and improving teaching outcomes.</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

Assessment of influencing factors of college and universities’ teaching effects using fuzzy and deep learning techniques

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月13日

查看全文
  • Zenghui He,
  • Xiaoyan Zhang,
  • Zhe Zhang,
  • Yi Lin,
  • Boen Yu

摘要

Effective teaching plays a vital role in universities and colleges, creating an excellent success rate for students and institutions. An effective teaching process creates a positive learning environment that helps students understand the material and improves their academic achievement. During the learning process, frequent assessments are necessary to ensure high-quality education, and the teaching process is evaluated based on various factors. However, the learning process presents challenges related to standardized test scores. Hence, there is a need for more advanced methodologies that can account for multiple factors and provide more precise, valuable assessments of instructional efficacy. This study aims to address the issue by employing fuzzy logic and deep learning methods to assess the factors influencing teaching effectiveness in higher education institutions. The research uses data from multiple institutions, accounting for variables such as the student-to-teacher ratio, student academic achievement, and teacher expertise. The data gathered is subjected to deep learning techniques to analyze the intricate correlations between the input factors and teaching success. The fuzzy logic technique is applied during the analysis to address the inherent uncertainty and vagueness in the data effectively. The student’s performance and teaching experience are used to evaluate the impact of teaching. The model achieved 97.89% accuracy, indicating it outperforms other models, such as TLDL, SEM-ANN, and regular ANN methods, in predicting teaching performance. It additionally achieved an accuracy of 86.6%, a recall of 75%, an F1-score of above 80%, and a low error rate of 10.4%. These performance measures demonstrate that the FDL technique is robust and dependable in handling data uncertainty and extracting valuable insights from complex educational datasets. The successful results indicate that combining fuzzy logic with deep learning significantly enhances prediction performance, providing educational institutions with a valuable tool for evaluating and improving teaching outcomes.