Um sistema gerador de jogos narrativos educativos utilizando Large Language Model

Abstract

The advancement of Artificial Intelligence (AI) has enabled new educational approaches, including the use of Large Language Models (LLMs) to create interactive educational content. This study presents the development of a narrative-based educational game generator that utilizes LLMs to automate the creation of interactive stories. The research explores the state of the art in educational games and interactive narratives, investigates prompt engineering techniques to optimize content generation, and implements a web-based platform integrating AI for dynamic and personalized storytelling. The developed tool allows users to select themes, customize characters, and interact with adaptable narratives, fostering active learning and engagement. Challenges related to narrative coherence, AI model limitations, and user interface adjustments were addressed to enhance usability. The results suggest that LLMs in education can facilitate the creation of interactive educational materials, reducing technical barriers and enabling innovative learning experiences.