Difficulty-aware Distractor Generation for Gap-Fill Items.
Chak Yan Yeung, John Lee, Benjamin K. Tsou · CityU Scholars · 2019
Many computer-assisted language learning (CALL) systems offer gap-fill items, often with multiple choices in order to facilitate immediate feedback. Automatic distractor generation can therefore be helpful in providing the multiple choices. While existing algorithms focus on proposing the most plausible distractors, many realistic scenarios make use of distractors at a variety of difficulty levels. This paper evaluates the use of a neural language model to rank distractors in terms of difficulty. Experiments show that BERT outperforms semantic similarity measures, in terms of both correlation to human judgment and classification accuracy of distractor plausibility. © 2019, Australasian Language Technology Association. All rights reserved.