Evaluation of a Template-Based Puzzle Generator for an Educational Programming Game

Authors

  • Yihuan Dong North Carolina State University
  • Tiffany Barnes North Carolina State University

DOI:

https://doi.org/10.1609/aiide.v13i1.12952

Keywords:

Educational games, puzzle, generation, procedural content generation

Abstract

Although there has been much work on procedural content generation for other game genres, very few researchers have tackled automated content generation for educational games. In this paper, we present a template-based, automatic puzzle generator for an educational puzzle programming game called BOTS. Two experts created their own new puzzles and evaluated generator-generated puzzles for meeting the educational goals, the structural and visual novelty. We show that our generator can generate puzzles with expert-designed educational goals while saving experts more than 80% of creation time, and these puzzles exhibit structural and visual novelty compared to expert-created puzzles. The contribution of this work is defined and implemented the first template-based automatic puzzle generator that saves expert time while incorporating expert-designed educational goals and enhancing puzzle.

Downloads

Published

2021-06-25

How to Cite

Dong, Y., & Barnes, T. (2021). Evaluation of a Template-Based Puzzle Generator for an Educational Programming Game. Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment, 13(1), 172-178. https://doi.org/10.1609/aiide.v13i1.12952