Paper: Similarity-Based Non-Scorable Response Detection for Automated Speech Scoring

ACL ID W14-1814
Title Similarity-Based Non-Scorable Response Detection for Automated Speech Scoring
Venue Innovative Use of NLP for Building Educational Applications
Session
Year 2014
Authors

This study provides a method that iden- tifies problematic responses which make automated speech scoring difficult. When automated scoring is used in the context of a high stakes language proficiency as- sessment, for which the scores are used to make consequential decisions, some test takers may have an incentive to try to game the system in order to artificially inflate their scores. Since many automated pro- ficiency scoring systems use fluency fea- tures such as speaking rate as one of the important features, students may engage in strategies designed to manipulate their speaking rate as measured by the system. In order to address this issue, we de- veloped a method which filters out non- scorable responses based on text similar- ity measures. Given a test response, the method generate...