Smart Automated Grading System Using Machine Learning Algorithm for Short Answers Questions

Abstract

An automated grading system was developed in this work to help teachers quickly grade multiple assignments at the same time, with fewer efforts, and especially with the Pandemic of COVID-19 and the challenges facing e-learning, in terms of correcting essay-answer tests. The use of automatic grading systems saves time and effort for educational institutions in achieving high-efficiency educational goals. considering that the number of students participating in online learning has increased dramatically, it is expected to reach more than 1.38 billion by 2023. Thus, it has become necessary to prepare a system for analysing and classifying students' responses to various types of electronic exams automatically. The k-Means method has been employed in many research fields. However, it has flaws due to the large number of parameters that affect its performance and the difficulty in managing the results. For that reason, the OPTICS clustering algorithm was employed in this work, as well as the African Buffalo Improvement technique, to get optimal results. The OPTICS algorithm clearly outperformed the K-Means algorithm, according to the results. The findings were greatly improved by using OPTICS and incorporating the African buffalo optimization technique.