Game-based Learning Model Using Fuzzy Cognitive Map

摘要

Fuzzy Cognitive Map (FCM) can be used to design game-based learning systems because it has the excellent ability of concept representation and reasoning. However, it can not (1) get new knowledge from existing data and (2) correct false transcendental knowledge by itself, which mars the game-based learning ability. This paper utilizes Hebbian learning rule to solve the first problem and uses unbalance degree to solve the second problem. As a result, the improved FCM has the ability of self-learning from both existing data and priori knowledge, and is more suitable for a game-based learning system. Based on the improved FCM, a novel game-based learning model is proposed, including a teacher submodel, a learner submodel and a set of learning mechanisms. The teacher submodel has the standard answers which can be deduced from the improved FCM. The learner submodel is bulit and adjusted according to the teacher's FCM, which reflects the learner's learning process. The learning mechanisms compute the difference between the outputs of the teacher submodel and the learner submodel, and control the whole game learning process according to the difference. Based on the proposed model, an automobile driving learning system is developed to prove the effectiveness of the proposed model. Extensive experimental results demonstrate our model validity in terms of controlling the learning process and the guiding learners learning.