PolyAQG Framework: Auto-generating assessment questions

Tan, Tee Hean * and Teh, Phoey Lee * and Zaharin Yusoff, (2021) PolyAQG Framework: Auto-generating assessment questions. In: 2021 IEEE Computing Conference on Computing (ICOCO 2021), 17-19 November 2021.

Preview

Text Teh Phoey Lee PolyAQG icoco 2021 polyaqg.pdf - Accepted Version Available under License Creative Commons Attribution Non-commercial. Download (612kB) | Preview

Abstract

Designing and setting assessment questions for examinations is always a necessary task for educators. In this article, we identify the research gaps in (semi-) automatically generating questions by evaluating all the available approaches developed thus far. We then propose a framework that puts together previous approaches and suggests ways to fill in their gaps. One hundred and thirteen pieces of literature relevant to question generation approaches have been reviewed and compared. For each of the approaches, the uniqueness of the techniques is explained. The PolyAQG Framework is presented with an explanation of how it would contribute to the solution of the problem, by improving the variety of the questions, increasing the total number of possible choices of question selections, as well as providing a better quality of questions. Apart from the framework, another novelty in this work is the innovative way a domain ontology is used to generate a wider variety of questions.

Actions (login required)

View Item