CoMeT: Integrating different levels of linguistic modeling for meaning assessment

摘要

This paper describes the CoMeT system, our contribution to the SemEval 2013 Task 7 challenge, focusing on the task of automatically assessing student answers to factual questions. CoMeT is based on a meta-classifier that uses the outputs of the sub-systems we developed: CoMiC, CoSeC, and three shallower bag approaches. We sketch the functionality of all sub-systems and evaluate their performance against the official test set of the challenge. CoMeT obtained the best result (73.1% accuracy) for the 3-way unseen answers in Beetle among all challenge participants. We also discuss possible improvements and directions for future research.