A Fuzzy Logic-Based Assessment System for Evaluating Student Performance in Open-Ended Tasks

Assessing student performance in open-ended tasks, such as essays, projects, and presentations, presents a significant challenge in educational settings. Traditional grading systems often fail to capture the nuances of these tasks, leading to inconsistencies and subjective evaluations. Fuzzy logic, with its ability to handle uncertainty and imprecision, provides a robust framework for assessing such tasks, offering a more flexible and comprehensive evaluation. This study proposes a Fuzzy Logic-Based Assessment System (FLBAS) designed to evaluate student performance in open-ended tasks. The system incorporates multiple criteria, including content quality, creativity, critical thinking, and presentation skills. Each criterion is evaluated using fuzzy sets, which allow for the representation of linguistic variables such as "excellent," "good," "average," and "poor." The fuzzy inference engine then combines these assessments to generate an overall performance score. The system is implemented using a Mamdani-type fuzzy inference model, which is widely recognized for its interpretability and effectiveness in educational assessments. The proposed FLBAS was tested on a dataset of student performances in a university course. The system's outputs were compared to traditional grading methods to evaluate its effectiveness and reliability. The results indicate that FLBAS provides a more nuanced and consistent evaluation, with reduced variability in scores compared to traditional methods. For example, in a sample task, the system rated 20% of the students as "excellent," 50% as "good," 20% as "average," and 10% as "poor." The traditional grading system, in contrast, showed greater variance, with some students receiving significantly different scores from different evaluators. The Fuzzy Logic-Based Assessment System offers a promising alternative to traditional grading methods for open-ended tasks. By incorporating fuzzy logic, the system provides a more flexible and accurate assessment of student performance, reducing subjectivity and enhancing consistency. The results suggest that FLBAS can be a valuable tool for educators seeking to improve the fairness and reliability of student evaluations in complex, open-ended tasks. Future work will focus on refining the system and exploring its application across different educational contexts.