[3H1-GS-10] AI application

[3H1-GS-10-02] Proposal of Neural Automated Essay Scoring Using Argument Mining

〇Misato Yamaura¹, Itsuki Fukuda¹, Masaki Uto¹ (1. The University of Electro-Communications)

Keywords:automated essay scoring, argument mining, deep learning

Automated essay scoring (AES) models based on deep neural networks (DNN) have recently achieved high accuracy. However, conventional neural AES models do not explicitly incorporate the logical structure of the essays. On the other hand, argument mining (AM) models that can estimate the logical structure in an essay have been proposed and enhanced its accuracy recently. This study therefore proposes a new DNN-AES model that integrates AM models to consider the logical structure of an essay, and demonstrate the effectiveness of the proposed method through experiments using benchmark data for AES.