Automated Topical Component Extraction Using Neural Network Attention Scores from Source-based Essay Scoring
This work addresses the need for formative feedback in essay revision by integrating AWE with neural AES, though it is incremental as it builds on existing methods for feature extraction.
The paper tackled the problem of linking automated writing evaluation (AWE) with neural automated essay scoring (AES) by extracting Topical Components (TCs) from source texts using attention layers, and found that performance in representing essays as rubric-based features and grading essays was comparable whether using automatically or manually constructed TCs.
While automated essay scoring (AES) can reliably grade essays at scale, automated writing evaluation (AWE) additionally provides formative feedback to guide essay revision. However, a neural AES typically does not provide useful feature representations for supporting AWE. This paper presents a method for linking AWE and neural AES, by extracting Topical Components (TCs) representing evidence from a source text using the intermediate output of attention layers. We evaluate performance using a feature-based AES requiring TCs. Results show that performance is comparable whether using automatically or manually constructed TCs for 1) representing essays as rubric-based features, 2) grading essays.