Automated Quiz Generation Tool Using Artificial Intelligence Techniques

Artificial intelligence has been used to construct numerous intelligent tutorial systems over the years with varying degrees of complexity and capability. They generally are constructed on top of a domain model of knowledge, some form of model of the student's learning, a pedagogical strategy that presents material to the student, and an interface. Automated question generation is often considered a subproblem within the larger tutorial domain (although not all automated question generation systems use artificial intelligence). Perhaps simpler to construct, the ability to generate questions to test a student is a very valuable tool. As with tutorial systems, question generation approaches vary widely. In recent years, approaches have concentrated on the use of large language models. Between tutorial system and question generation attempts, research indicates a significant investment in time and effort to construct such systems. This paper presents an AI-based approach to automated question generation that foregoes the language model/language interface and uses a much shallower domain model to simplify construction, yet provides beneficial results to the student. This paper introduces the QuizGenie system to demonstrate the approach taken. The work is still in progress but shows promise as a study resource for students.