iGAME: Cognitive Game Analysis Through Eye Movements of the Player

The main goal of this work is to use eye movement tracking in classifying players based on their gameplay data and analyze their experiences while playing Tic Tac Toe. We collected data by tracking eye gaze and mouse movement of 20 participants while they played the game. The collected data was pre-processed and cleaned to prepare it for classification. We used six classification algorithms, including SVM, Naive Bayes, KNN, Decision Tree Classifier, Random Forest, and XG Boost, to classify players into beginner, intermediate, and expert levels. The accuracy of the classification algorithms ranged from 94.09% to 95.51%. We further evaluated the performance of each algorithm by generating classification reports. Our analysis of the collected data allowed us to gain valuable insights into how different levels of players experience Tic Tac Toe and which aspects of the game may be more challenging for certain skill levels. This information can be used to inform game design and optimize the gameplay experience for all players. The results of this study have the potential to enhance player engagement.