Automated essay grading is one of the most important educational applications of NLP. Given an essay, an automated essay grader's goal is to provide annotations with respect to various dimensions of essay quality such as coherence, argument strength, and technical quality, all without human intervention. These annotations can take forms including scores (e.g. on a scale of 1.0 to 4.0, how clear is the essay's thesis?) and error labels (e.g. is any part of the essay phrased in a confusing way?). Being able to automatically generate these essay annotations is important because it can help students improve their writing by providing immediate feedback that would otherwise only be available through a teacher's active supervision.
This page serves as a distribution site for information relating to our work advancing the state of the art in several automated essay grading tasks. It includes our published works on the tasks and corpus annotations: