Answers are grounded in this paper and cite their sources.
Essay classification is a crucially important task, considering the increasing attention towards AI tools such as ChatGPT, and the raised concerns with regard to academic integrity. Existing text classification approaches, including neural network-based methods and feature-based methods, are often unable to effectively analyze diverse types of information in a combined way. Moreover, they are mostly tailored for English data, and focus on addressing specific topics, such as fake news, hate speech, and misinformation. On the other hand, machine learning research conducted on essay data has been limited to automated grading and off-topic detection tasks. In this paper, we propose a deep neural network-based model that combines text modeling and linguistic features via data fusion, with the aim to provide robust classification capabilities. Our experiments with datasets containing short essays in English and Spanish written by L2 learners, show that it is possible to accurately perform essay classification, and that our model is competitive, even when compared to popular feature-based and neural network-based approaches.
No takes yet. Share an insight, caveat, or question.
Corizzo et al. (2023) studied this question.