Blog: Large Language Models for Curriculum Mapping with a Focus on Employable Skills

In the era of rapid technological evolution, where required employable skills are expected to change rapidly, higher education institutions (HEIs) need to prepare students for an uncertain and dynamic job market. Large Language Models (LLMs) - tools like GPT-4 - are emerging as a promising way to tackle this challenge by assisting with curriculum mapping and design.  

Curriculum mapping is the process of aligning course descriptions, learning activities, and assessments with the skills needed for employment. Traditionally, experts carry out this process manually. However, this manual approach requires considerable resources and faces several difficulties. For instance, experts may interpret employable skills differently, and human error or bias can occur. These issues make it harder to update curriculum maps frequently, even though the job market is rapidly changing. 

Recent studies have investigated whether large language models (LLMs) can automate this process [1]. Their findings suggest that LLMs help speed up curriculum mapping, but sometimes, they are less accurate than human experts. Despite this limitation, a promising strategy is to combine LLMs with Retrieval Augmented Generation (RAG) and then have experts review the results. This hybrid method could make curriculum mapping more efficient while maintaining high accuracy standards. 

Overall, integrating LLMs into curriculum mapping can offer several advantages:

• Accelerating the Mapping Process: LLMs can quickly examine course descriptions, topics, and learning outcomes, aligning them with the employability skills needed for modern jobs. This allows for faster identification of curriculum gaps and opportunities for improvement. 
• Quality Assurance Support: By comparing course content to the required skills, LLMs act as an additional quality check. They can pinpoint overlapping material or missing topics and suggest ways to enhance the curriculum’s quality. 
• Analyzing Curricula-Related Feedback: Universities gather extensive feedback and performance data, such as course evaluations, student surveys, and learning analytics. LLMs can be used to review and analyze these data sets to uncover patterns and insights that can guide curriculum development. 
• Curriculum Design Assistance: Beyond mapping and feedback analysis, LLMs can also help create new curricula. Given established educational standards and desired employability skills, LLMs can propose course outlines, content topics, and assessment methods as a starting point for further refinement by curriculum developers. 

Despite their potential advantages, LLMs also pose certain challenges. They may have difficulty interpreting complex qualities like lifelong learning and ethical responsibility, creating obstacles when mapping curricula. Furthermore, since LLMs rely on training data, they can reflect biases that influence their recommendations. Another issue arises from inconsistent definitions of employable skills. For instance, “communication” can include public speaking, writing, or negotiation, and it can be hard for LLMs to determine which specific aspect of “communication” best aligns with a given curriculum. 

To summarize, incorporating large language models (LLMs) into the curriculum mapping process offers a transformative avenue for enhancing students’ employability skills. LLMs can notably streamline mapping procedures and inform curriculum development and design by expediting the analysis and synthesis of curriculum-related feedback. Nevertheless, it is crucial to consider risks such as misinterpretation or overreliance on specific data while utilizing LLMs for this purpose. Adopting advanced techniques—such as reinforcement learning and knowledge graphs—holds promise for further improving LLM performance in this area. 

Reference:
[1] Mapping Employable Skills in Higher Education Curriculum Using LLMs | SpringerLink 

 

- Written by Hamayoon Behmanush