Paper: Automatic Generation of Challenging Distractors Using Context-Sensitive Inference Rules

ACL ID W14-1817
Title Automatic Generation of Challenging Distractors Using Context-Sensitive Inference Rules
Venue Innovative Use of NLP for Building Educational Applications
Session
Year 2014
Authors

Automatically generating challenging dis- tractors for multiple-choice gap-fill items is still an unsolved problem. We propose to employ context-sensitive lexical infer- ence rules in order to generate distractors that are semantically similar to the gap tar- get word in some sense, but not in the par- ticular sense induced by the gap-fill con- text. We hypothesize that such distrac- tors should be particularly hard to distin- guish from the correct answer. We focus on verbs as they are especially difficult to master for language learners and find that our approach is quite effective. In our test set of 20 items, our proposed method de- creases the number of invalid distractors in 90% of the cases, and fully eliminates all of them in 65%. Further analysis on that dataset does not support o...