Multi-Criteria Decision-Making for Evaluation of Student Academic Performance Based on Objective Weights

Abstract

Student academic evaluation is part of the learning process in order to control the student's learning progress. The evaluation will show whether the student will pass or fail and for an instructor to guide for future evaluations on performance. There are criteria such as student's gender, student's age when they are registered in university, student's 1st semester GPA, etc., which exist in academic data can be utilized to get the student academic performance using multicriteria decision making. Multi-Objective Optimization by Ratio Analysis (MOORA) and Simple Multi-Attribute Rating (SMART) was two simple technique in multi-criteria decision making that the criteria weight can be determined objectively using entropy and gain values. This paper tries to evaluate the student academic performance using MOORA and SMART with criteria weight and sub-criteria weight resulted from entropy and gain. Decision output out of MOORA and SMART then compared with actual data using confusion matrix to discover the performance of those criteria and sub-criteria weight. The result showed that the performance of criteria weight with accuracy was 60.9 percent and the criteria of fourth-grade point average have the biggest impact on student academic evaluation with 0.1589 of weight. The result of this research can be used to help the instructor to determine the weight of student criteria for future recommendations and evaluations on student performance.