A Study of Correcting Handwritten Answers for Short Essay Self-learning Systems
Takahiro Yamasaki, Ayako Hiramatsu · 2023
We aim to develop an e-learning system for improving Japanese language proficiency. This system not only provides immediate scoring and advice for learners’ answers but also automatically generates questions to offer a wide range of problems across various fields. As the first step for this purpose, we aim to accurately extract the content described in handwritten essays. When dealing with handwritten characters, OCR software alone cannot achieve sufficient accuracy due to noise, distortion, and idiosyncratic handwriting. Therefore, we first train a neural network for character recognition using a handwritten character database, and then divide the essay images into single characters to determine the most likely character candidates. Furthermore, we treat character identification as a fill-in-the-blank problem for sentences and use BERT’s Masked Language Model task to determine characters that form natural sentences. Applying this method to actual handwritten essays on manuscript paper, we were able to extract characters with higher accuracy than before.