A MAMDANI-TYPE FUZZY INFERENCE SYSTEM TO AUTOMATICALLY ASSESS DIJKSTRA'S ALGORITHM SIMULATION
M. Gloria Sánchez-Torrubia, Carmen Torres-Blanc · 2010
In education it is very important for both users and teachers to know how much the student has learned. To accomplish this task, GRAPHs (the eMathTeacher-compliant tool that will be used to simulate Dijkstra's algorithm) generates an interaction log that will be used to assess the student's learning outcomes. This poses an additional problem: the assessment of the interactions between the user and the machine is a time- consuming and tiresome task, as it involves processing a lot of data. Additionally, one of the most useful features for a learner is the immediacy provided by an automatic assessment. On the other hand, a sound assessment of learning cannot be confined to merely counting the errors; it should also take into account their type. In this sense, fuzzy reasoning offers a simple and versatile tool for simulating the expert teacher's knowledge. This paper presents the design and implementation of three fuzzy inference systems (FIS) based on Mamdani's method for automatically assessing Dijkstra's algorithm learning by processing the interaction log provided by GRAPHs.