AutoTag & TagMap: LLM-Powered Moodle Plugins for Pedagogical Alignment Checks

Pedagogical alignment denotes the coordination between learning outcomes, teaching and learning activities and assessment tasks. For any learning outcome of a course, teachers should design corresponding learning activities and consequently assess its achievement. Fair and valid assessments require a proper pedagogical alignment. At a low-level, this can be verified by tagging learning material, e.g., lecture notes or presentations, and assessment items, e.g., quiz questions, and checking a homogeneous coverage. Learning Management Systems (LMS) such as Moodle enable teachers to tag both resources and questions. To ease the tagging, named entity recognition technologies can be used. With the advent of Large Language Models (LLMs) such as GPT-4, this task has seen a new momentum. In this paper, we present AutoTag and TagMap, two Moodle plugins which will help teachers to check the pedagogical alignment of their online courses. The first plugin, AutoTag, leverages GPT-4 to provide automatic resource tagging support to teachers. An existing plugin for question generation also implements this logic. The second plugin, TagMap, independent of the other, visualizes concept coverage in both resources and questions to help teachers in verifying the pedagogical alignment and identifying possible shortcomings. We exemplify the usage of both plugins in a nephrology and urology course.