A Graph-based Approach for Adaptive Serious Games

In higher education, researchers have recently focused on exploring automated, personalized instructional systems to enhance students’ learning experiences. Motivated by this, we propose a personalized instructional system using a straightforward graph system to offer an educational game that is effective for students and intuitive for developers. Our system uses a directional graph, called an action graph, for representing solutions to in-game problems based on possible player actions. Through our proposed algorithm, a serious game integrated with our system would both detect player errors and provide personalized assistance to direct a player in the direction of a correct solution. To verify system performance, we present comparison testing on a group of students engaging in the game both with and without AI. Students who played the AI-assisted game showed an average 20% decrease in time needed and average 58% decrease in actions taken to complete the game.