Predicting Desirable Revisions of Evidence and Reasoning in Argumentative Writing
This work addresses the challenge of providing automated feedback for improving argumentative writing skills in students, though it appears incremental as it builds on existing classification methods.
The paper tackled the problem of classifying desirable evidence and reasoning revisions in student argumentative writing, finding that models using context information, either alone or with feedback, were most successful in identifying these revisions.
We develop models to classify desirable evidence and desirable reasoning revisions in student argumentative writing. We explore two ways to improve classifier performance - using the essay context of the revision, and using the feedback students received before the revision. We perform both intrinsic and extrinsic evaluation for each of our models and report a qualitative analysis. Our results show that while a model using feedback information improves over a baseline model, models utilizing context - either alone or with feedback - are the most successful in identifying desirable revisions.