Create Dynamically Adaptive Test on the Fly Using Fuzzy Logic
Konstantina Chrysafiadi, Maria K. Virvou · 2018
Past decade digital education has become a burning issue. Furthermore, there is a growing need of providing personalized teaching and learning processes. This is achieved through adaptive learning systems and adaptive educational applications, which enrich the tutoring and learning processes with “intelligence”. In this way, they adapt either the learning material or the tutoring and learning processes to each individual student's needs and abilities, offering him/ her a personalized learning experience. A significant part of the learning material is testing. Tests have to be adapted to the learners' particular characteristics, needs and abilities, in order to be more effective and accurate. In this paper, an algorithm that creates adaptive tests on the fly is presented. In particular, the presented algorithm is based on fuzzy logic and takes into consideration the knowledge dependencies that may exist among the different domain concepts of the learning material. It decides about the test items that have to be included in the created test on the fly, while the learner is solving the test, and tries to detect either the learner's deficiencies in knowledge-depended concepts that have been taught previously or the following domain concepts that the learner knows partially or completely. The particular algorithm has been applied for the creation of tests in a web-based educational environment for the teaching of computer programing and the programming language `C', which has been used by the students of a postgraduate program in Informatics, at the University of Piraeus. The aforementioned algorithm has been evaluated and the evaluation results are very encouraging.