Classy Trash Monster: An Educational Game for Teaching Machine Learning to Non-major Students

Joonhyung Bae, Karam Eum, Haram Kwon, Seolhee Lee, Juhan Nam, Young Yim Doh

Abstract: As machine learning (ML) became more relevant to our lives, ML education for college students without technical background arose important. However, not many educational games designed to suit challenges they experience exist. We introduce an educational game Classy Trash Monster (CTM), designed to better educate ML and data dependency to non-major students who learn ML for the first time. The player can easily learn to train a classification model and solve tasks by engaging in simple game activities designed according to an ML pipeline. Simple controls, positive rewards, and clear audiovisual feedback makes game easy to play even for novice players. The playtest result showed that players were able to learn basic ML concepts and how data can impact model results, and that the game made ML feel less difficult and more relevant. However, proper debriefing session seems crucial to prevent misinterpretations that may occur in the learning process.