Implementation of Handwriting Recognition and Answer Evaluation with Recurrent Neural Network

Abstract

Many teachers “still utilize exams to gauge their pupils' knowledge. However, as the teacher must comprehend each answer provided by each student individually, analyzing the exam response can take a long time. Additionally, different handwriting styles may lead to errors in scoring the response, resulting in an erroneous mark. This study suggests a web-based program that can read students' handwriting to assist teachers in assessing students’ responses. The objective of this study was to develop a web-based application that can accept student responses as input and output correctness.