Using Large Language Models for Automatic Item Generation to Create Multiple-Choice Questions

• In Event: GSIC Poster Session II

Abstract

This research-in-progress investigates an automatic item generation (AIG) methodology which utilizes subject matter expert (SME) model and large language model capabilities to create a large number of high quality items. We follow Gierl et al. (2017) three-step method in AIG. Items will be evaluated by textual features and by SME.

Authors

• Nour Armoush, Unviersity of Alberta

• Mark J Gierl, University of Alberta