Learner Modeling in Educational Games Based on Fuzzy Logic and Gameplay Data

Abstract

Over the last years there has been a growing interest in the use of educational games as learning tools. Educational games have proven to contribute in enhancing student motivation, increasing their engagement and providing them with personalized and adaptive learning. Learner modeling is a prerequisite when it comes to adaptive learning; it is used to represent student's knowledge, needs, and characteristics. This paper presents a modeling technique based on fuzzy logic that uses gameplay data and expert rules to predict learners preferred learning and playing styles. To test the fuzzy rule-based systems, the educational game Woodland was designed bearing in mind the VARK learning styles and the Bartle playing styles. High school students played the educational game Woodland and the results of the FRBSs were compared with the result of the questionnaires. A great correlation was found between the FRBSs results and the questionnaire results.