Automatic Multiple-Choice and Fill-in-the-Blank Question Generation from Arbitrary Text

Abstract

In this paper, we propose three novel algorithms to help educators generate questions to evaluate learners’ comprehension of the learning material. First, we propose the Automatic Question Generator algorithm that automatically generates multiple-choice, true-or-false, and fill-in-the-blank questions from any arbitrary text. Second, we propose the Automatic Distractors Generator algorithm that automatically generates multiple wrong but relevant answers for the multiple-choice questions. Third, we propose the Automatic Answers Generator algorithm that automatically generates multiple synonyms to the correct answer of the fill-in-the-blank questions to be accepted as correct answers. The algorithms have been used on finance and business learning material as case studies for evaluation.