An Automated Writing Evaluation System for Supporting Self-monitored Revising

Diane J. Litman, Tazin Afrin, Omid Kashefi, Christopher Olshefski, Amanda Godley, Rebecca Hwa

Abstract: This paper presents the design and evaluation of an automated writing evaluation system that integrates natural language processing (NLP) and user interface design to support students in an important writing skill, namely, self-monitored revising. Results from a classroom deployment suggest that NLP can accurately analyze where and what kind of revisions students make across paper drafts, that students engage in self-monitored revising, and that the interfaces for visualizing the NLP results are perceived by students to be useful.