Design Specifications for an Interactive Teaching Tool for Game AI using Gomoku

Sammanfattning: This thesis seeks to understand and improve how students can learn the fundamentals of strategic game AI using a game-like application. The work focuses on the design specifications of a mockup application that can be used to teach a user the concepts behind the Minimax and Alpha-Beta Pruning algorithms using the strategic game Gomoku. The work aims to shed light on how technology can be used to teach students about technology and explore the possible ways it can be facilitated engagingly. The design of the mockup is based on concepts from education, human-computer interaction (HCI), and strategic game AI. The experimental learning model developed by David Kolb was used to structure the learning content, while frameworks from HCI were used to analyze the target audience and design the interface. The primary focus of the design was to rationalize the logic of the AI to the human audience using scenarios and exercises. The mockup was evaluated using a cognitive walkthrough with 5 participants. The results of the study indicate that the design can deliver an effective learning tool for teaching how game trees and the Minimax algorithm work. However, the results also suggest that the application struggles to teach the user more about the complex Alpha-Beta Pruning algorithm. Despite these results, more research and user tests are needed to determine whether the application is effective for the target audience.