Scholarlink

<p>Aligning course outcomes with national qualification framework standards and descriptors is a critical step in ensuring educational quality, relevance and international compliance. The National Qualification Framework (NQF) categorizes learning outcomes into three primary dimensions: Knowledge, Skills, and Values / Competencies. Manual classification of course outcomes into these descriptor categories is a time-consuming and error-prone process due to its subjectivity. This article presents a new approach that leverages Natural Language Processing (NLP) techniques, including WORD2VEC (W2V) and Term Frequency–Inverse Document Frequency (TF-IDF), combined with a proposed deep learning model, to automate this classification and mapping process. Our approach incorporates Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs) to effectively classify course outcomes by leveraging external knowledge encoded in pre-trained word embeddings. The integration of pre-trained word embeddings enhances the model’s ability to <i>understand</i> context and slight differences in course descriptions. Experimental results demonstrate that even relatively simple deep learning architectures, when paired with rich embeddings such as WORD2VEC or TF-IDF, can achieve high classification accuracy. In conclusion, this research offers a practical and scalable solution to support educators and curriculum designers and provides valuable insights for future applications in curriculum development and internationalization of qualifications.</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

NQF: A deep learning model for classifying course outcomes in the national qualification framework

  • 期刊论文

  • 发表日期：

    2025年6月13日

查看全文
  • Ammar El-Hassan,
  • Mohammad Azzeh,
  • Bashar El-Rashdan

摘要

Aligning course outcomes with national qualification framework standards and descriptors is a critical step in ensuring educational quality, relevance and international compliance. The National Qualification Framework (NQF) categorizes learning outcomes into three primary dimensions: Knowledge, Skills, and Values / Competencies. Manual classification of course outcomes into these descriptor categories is a time-consuming and error-prone process due to its subjectivity. This article presents a new approach that leverages Natural Language Processing (NLP) techniques, including WORD2VEC (W2V) and Term Frequency–Inverse Document Frequency (TF-IDF), combined with a proposed deep learning model, to automate this classification and mapping process. Our approach incorporates Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs) to effectively classify course outcomes by leveraging external knowledge encoded in pre-trained word embeddings. The integration of pre-trained word embeddings enhances the model’s ability to understand context and slight differences in course descriptions. Experimental results demonstrate that even relatively simple deep learning architectures, when paired with rich embeddings such as WORD2VEC or TF-IDF, can achieve high classification accuracy. In conclusion, this research offers a practical and scalable solution to support educators and curriculum designers and provides valuable insights for future applications in curriculum development and internationalization of qualifications.