Software Design of Autocorrection Essays on the Website and Application

Abstract

Traditional essay assessment methods are often time-consuming and prone to subjectivity. This study proposes a novel Automated Essay Scoring (AES) system, "Essay Mathematic Auto Correction (Emath Toco)," featuring web and mobile app interfaces. Emath Toco leverages visual stimuli and deep learning algorithms like 1D CNN, NasNet Mobile, and GoogleNet to offer objective and efficient essay evaluation. Extensive testing on a 40/60 training/testing data split yielded accurate data classification, validating successful implementation on Flutter-built Android applications and a Firebase-powered web interface. User experience surveys revealed positive feedback on Emath Toco's ease of use, visually appealing interfaces, and effective data collection, confirming its user-friendliness. Emath Toco's innovative use of visual stimuli and deep learning algorithms significantly reduces subjectivity and improves the accuracy of essay evaluation. Emath toco is promising technology with the potential to revolutionize essay assessment and educational methodologies. The research contributes to the field of automated essay scoring in two key ways. First, by integrating visual stimuli as a novel approach, Emath Toco expands the range of factors considered in scoring, potentially leading to more comprehensive and efficient. Second, the successful implementation of the system on both web and mobile platforms demonstrates its flexibility and accessibility, offering educators a versatile tool regardless of technological limitations.