Published Paper Details:

This study presents a comparative analysis of fuzzy logic and traditional statistical methods for evaluating academic performance, focusing on the application of fuzzy membership functions. In traditional methods, such as the use of weighted averages, student performance is assessed based on fixed boundaries, often leading to rigid and sometimes unrealistic evaluations. In contrast, fuzzy logic provides a flexible framework that allows for the modeling of uncertainty and partial truth by employing membership functions that categorize performance levels like "Poor," "Average," "Good," and "Excellent" with varying degrees of membership. Using a case study of students' test and project scores, this research demonstrates how fuzzy logic-based evaluations can offer more nuanced and context-sensitive assessments compared to conventional statistical approaches. By analyzing the impact of different fuzzy rules and membership function shapes on the overall performance scores, the study highlights the advantages of fuzzy logic in capturing the complexities of academic performance. The results show that fuzzy logic provides a more adaptable and informative evaluation, especially when dealing with ambiguous or borderline cases, thereby offering a more realistic alternative to traditional methods.