A Fuzzy Logic Approach for Student Performance and Exam Integrity Assessment in a Web-Based Smart Quiz System in Higher Education
Purpose – This research aims to develop a Web-Based Smart Quiz System that integrates performance analytics and user behavior monitoring to support more objective and comprehensive learning evaluations. The problem raised is the limitations of the online quiz system in combining fuzzy logic-based performance analysis with exam integrity supervision on the system dashboard. Design/methods/approach – This system is developed using a prototype model. Student performance and exam integrity were analyzed using a Mamdani fuzzy logic approach based on four input variables, namely score percent, integrity, attempt count, and time ratio. For fuzzy inference processing, warning count, tab switch count and fullscreen exit count were aggregated into a single integrity violations index, which was used as the integrity variable. The system is also equipped with examination integrity features, an analytics dashboard, and question item analytics to support adaptive evaluation and examination integrity assessment. Findings – The results showed that 81 quiz attempts were classified into 33 Superior, 39 Good, 7 Enough, and 2 Needs Coaching. Risk analysis showed that 79 attempts were in the Safe category, while 2 attempts required further attention. These results indicate that the proposed system can support adaptive performance evaluation while facilitating examination integrity monitoring during online assessment. Research implications/limitations – The study was conducted within a single higher education institution and relied on predefined membership functions and rule bases. Future studies should involve larger and more diverse populations and explore longitudinal validation of the model. Originality/value – This study presents a prototype-based integrated assessment platform that combines fuzzy performance evaluation, examination integrity monitoring, user behaviour analytics, and question item analytics within a single web-based Smart Quiz System. The proposed platform is intended as a decision-support tool for learning evaluation and provides a foundation for future validation and enhancement using larger datasets and advanced analytics methods.