An Overview of Early Prediction and Classification Models for the Performance Analysis of Degree Students

The students who are working to earn a bachelor's degree at a college or university are known as degree students. The success of the degree students plays an essential role in enhancing the values of the college; also, the performance of the institution is represented by the student?s success rate. Hence, in the education field, significant attention has been gained by the prediction of degree student?s Academic Performance (AP). The students? performances are assessed by formal exams, quizzes, research projects, case studies, and homework and the results are represented in a Grade Point Average (GPA) format. The performances of the students are influenced by students? e - learning activity, family support, student internal assessment grade, previous assessment grade, low entry grades, accommodation, and GPA. For providing adequate support to enhance the student?s performance, an effective as well as early prediction of the student?s performance is required. To forecast the students? performances in the academic period, several prediction and classification models were developed by the prevailing frameworks. Hence, this review analyzes the developed prediction and classification models utilized for the early performance prediction of the students. The evaluation of the developed models exhibited the Machine Learning (ML) and Deep Learning (DL) methods? superiority in performance prediction by attaining effective prediction accuracy, precision, and recall.