NC State enhances educational games with real-time data mining

Researchers at NC State University are using data mining to improve educational games. Their aim is to help students develop collaborative problem-solving skills in real time. The team is creating software tools that analyze how effectively students learn while playing. These tools use algorithms to assess and predict student learning outcomes, making games more responsive to individual needs. Wookhee Min, a senior research scientist, noted that the tools can adapt as more data is collected from students' gameplay. The research focuses on measuring cognitive aspects of collaborative problem-solving without disrupting gameplay. This approach allows the software to identify patterns in student behavior using data-driven methods. In a recent study, 61 middle school students played an educational game called EcoJourneys. This game emphasizes teamwork to solve a problem involving sick fish on a local farm. The game includes various phases where students investigate and deduce answers together. Researchers analyzed gameplay data to understand better how effective the game is for teaching collaborative skills. They found that the software could also modify the game in real time based on student choices. For instance, if students struggle with certain concepts, the game can adjust to reinforce those ideas. The findings were shared at a learning analytics conference in Dublin earlier this month. The research has financial support from the National Science Foundation, and several NC State researchers contributed to the study.