Hybrid learning style identification and developing adaptive problem-solving learning activities

• We identified the multiple learning styles that affect the learning performance in a game-based problem-solving activity.
• The system had good performance and the technology was accepted (e.g., “perceived ease of use” and “perceived usefulness”).
• Data mining techniques were employed in clustering the learning styles.

Learning style refers to an individual’s approach to learning based on his or her preferences, strengths, and weaknesses. Problem solving is considered an essential cognitive activity wherein people are required to understand a problem, apply their knowledge, and monitor behavior to solve the issue. Problem solving has recently gained attention in education research, as it is considered an essential ability for effective learning. This study aims to investigate the relationship between learning styles and learning performance. To provide adaptive suggestions for optimizing problem-solving abilities, developed a hybrid learning style identification (HLSI) mechanism based on a k-means clustering algorithm was developed. The participants were 67 undergraduate students. The experiment demonstrated that HLSI can successfully cluster learning styles into three or four combinations based on learning performance, which suggests that the data mining technique can successfully explore multiple learning styles in problem-solving abilities. Additionally, 13 teachers were included in the study to discuss the effectiveness of the HLSI mechanism, and the results indicated a 95% probability of obtaining an above-average acceptance of the proposed system.