Investigating Methods for Identifying Beginner Player Mistakes in Incomplete Information Games: An Analysis Based on Pokémon Battles

Abstract

In today's rapidly changing era, it is important to shift from passive learning to active learning in the field of education in order to be creative and active. To achieve active learning, it is important for students to "learn proactively, interactively, and deeply." To achieve this, gamification-based education has been proposed, and it has been suggested that a condition for this is that players need to "think carefully about each move." In this study, we focus on "pre-optimization" and "room for choice" in incomplete information games, with the elements that encourage contemplation as the construction of strategies. Therefore, we take up Pokémon, which meets these conditions, as a research material and aim to "construct a playing skill improvement support system that interactively supports the construction of strategies appropriate for players" to realize "proactive, interactive, and deep learning." In this paper, as a preliminary step in the research, we propose a method to discover bad moves made by players who need guidance using the structured knowledge of experts.