Towards Process-Oriented, Modular, and Versatile Question Generation that Meets Educational Needs

Abstract

NLP-powered automatic question generation (QG) techniques carry great pedagogical potential of saving educators' time and benefiting student learning. Yet, QG systems have not been widely adopted in classrooms to date. In this work, we aim to pinpoint key impediments and investigate how to improve the usability of automatic QG techniques for educational purposes by understanding how instructors construct questions and identifying touch points to enhance the underlying NLP models. We perform an in-depth need finding study with 11 instructors across 7 different universities, and summarize their thought processes and needs when creating questions. While instructors show great interests in using NLP systems to support question design, none of them has used such tools in practice. They resort to multiple sources of information, ranging from domain knowledge to students' misconceptions, all of which missing from today's QG systems. We argue that building effective human-NLP collaborative QG systems that emphasize instructor control and explainability is imperative for real-world adoption. We call for QG systems to provide process-oriented support, use modular design, and handle diverse sources of input.

Surgment: Segmentation-enabled Semantic Search and Creation of Visual Question and Feedback to Support Video-Based Surgery Learning

“…Techniques to generate visual questions that target higher-order thinking skills are limited. Recent work on question generation [51,87] also emphasized the importance of incorporating expert input to ensure the quality of the produced questions. Surgment offers another example of the importance of involving human experts at all stages in the question-creation process.…”

Evaluating the Knowledge Dependency of Questions

“…MCQ makes student assessment feasible, as it disambiguates a question by providing options to choose from (Rachmat and Arfiandhani, 2019). Especially, there have been multiple findings that active learning through question answering helps increasing learning gain of students (Crouch and Mazur, 2001;Koedinger et al, 1997;Wang et al, 2022). However, instructors suffer from generating high quality MCQs due to their limited resources.…”

“…Also, the suggested answerability does not account for the source text given with the MCQ. (Wang et al, 2022) points out the low adoption of QG Systems in classrooms, requesting the QG system researchers to focus on educational needs.…”