Data Augmentation for Automated Essay Scoring using Transformer Models

Transfer learning is proving quite useful in Natural Language Processing. One of the most important problem in Natural Language Processing is Automated essay scoring, which remains partially unsolved especially when we are dealing with single language model capable to evaluate essays of multiple topics. In this work we examine the effectiveness of transformer models like BERT, RoBERTa, etc using data augmentation technique to build a model capable of evaluating essays of multiple topics having little data.