PREDICTION OF STUDENTS’ ACADEMIC PERFORMANCE USING MULTIPLE LINEAR REGRESSION

Abstract

Nowadays, the education system has been changed from teacher centered approach to learner-centered approach. Thus, students’ related features need to analyze to predict students’ academic performance to provide the active learning in educational system. In this system, Multiple Linear Regression (MLR) is applied to predict the students’ academic performance on the UCI Machine Learning Repository’s student performance data set. The student’s academic performance prediction model is built by using Multiple Linear Regression method. The purpose of this system is to predict the students’ final grade based on previous grades and relevant features. Feature selection method is used to reduce the number of input variables that are believed to be the most useful to a regression model in order to predict the target variable. For evaluating the result of prediction model performance applied on students’ academic performance model use four different measures: accuracy, precision, recall and f-measure. This system is implemented using C# programming language with Microsoft Visual Studio IDE and Microsoft SQL Server as Database Engine.