Javascript must be enabled for the correct page display

Designing Game-Based Machine Learning Education for Dutch High School Students

McGinley, Tex (2024) Designing Game-Based Machine Learning Education for Dutch High School Students. Bachelor's Thesis, Artificial Intelligence.

[img]
Preview
Text
bAI2024TexMcGinley.pdf

Download (7MB) | Preview
[img] Text
toestemming.pdf
Restricted to Registered users only

Download (524kB)

Abstract

This paper explores the use of game-based learning (GBL) to teach machine learning (ML) concepts, specifically recommender systems, to Dutch high school students in the VWO educational stream. The study addresses the gap in AI education within the Dutch curriculum by developing an engaging and accessible educational program. This program includes a comprehensive teaching guide, presentation slides, and the Movie Ranker game, which simulates a Netflix-style recommendation system to illustrate ML principles. The program was created using the preparation, design, and improvement methodology adapted from Park and Kwon’s South Korean paper, with the goal of demystifying AI, educating about recommender systems, introducing reinforcement learning (RL) concepts through the Multi-Armed Bandit (MAB) problem, and highlighting the ethical considerations and implications of these systems. The created Movie Ranker game incorporates a competitive mechanic through the use of an ε-Greedy bandit, which the player must compete against. The results of the implemented ε-Greedy algorithm showed that it is effective in the created environment, performing better than a random agent in a statistically significant manner. This competitive element not only engages students but also helps them understand RL concepts like the exploration vs. exploitation dilemma. While the program has yet to be tested and evaluated in a real-world classroom setting, preliminary findings suggest that the game’s interactiv

Item Type: Thesis (Bachelor's Thesis)
Supervisor name: Cardenas Cartagena, J. D.
Degree programme: Artificial Intelligence
Thesis type: Bachelor's Thesis
Language: English
Date Deposited: 30 Jul 2024 10:01
Last Modified: 30 Jul 2024 10:01
URI: https://fse.studenttheses.ub.rug.nl/id/eprint/33548

Actions (login required)

View Item View Item