A Bibliometrics-Based Systematic Review on Automated Essay Scoring in Education

Advances in information technology and computer science have brought significant changes to various sectors of human life. In education, computers have helped ease a variety of human tasks, especially for large-scale and repetitive tasks such as essay grading. Current's computer capabilities make it possible to do essay scoring automatically. Research related to automatic essay scoring continues to develop by trying certain techniques and methods to improve the accuracy of automatic scoring close to that of human scoring. This systematic review aims to explore the 21st century applications of automated essay grading used in education. We carried out a bibliometric analysis of the meta data obtained from the SCOPUS database using the Bibliometrix and VOSviewer software reporting the main themes of implementing automated essay scoring applications in educational studies, providing an overview of the current state of the art and future directions of research and applications. Based on bibliometric analysis, we found that research related to automatic essay assessment in education has a fluctuating development trend and the current direction of AES research has used deep learning techniques such as long short-term memory, transfer learning and bert to build a better AES.