Using Gaussian membership functions for improving the reliability and robustness of students’ evaluation systems

In this paper, a more reliable system of student evaluation based on gaussian membership functions will be introduced. The proposed method is modeled as a three fuzzy nodes system. Each of the three nodes applies fuzzification, fuzzy inference, and defuzzification in considering the difficulty, importance and complexity of questions. The first node computes the difficulty of questions as a function of the fuzzified average accuracy and average time rates of questions. The second node computes the cost of answering questions as a function of its difficulty and complexity. The third node computes the degree of adjustment required by questions as a function of its answer-cost and importance. The accuracy and answer-time rates of questions are obtained from students’ answerscripts while the complexity and importance of questions are obtained by a domain expert, i.e., teachers and/or examiners. In order to improve the reliability and robustness of the system, Gaussian membership functions (MFs) are proposed as an alternative to the traditional triangular MFs.