Banca de DEFESA: RAIMUNDO NONATO BEZERRA NETO

Uma banca de DEFESA de DOUTORADO foi cadastrada pelo programa.
STUDENT : RAIMUNDO NONATO BEZERRA NETO
DATE: 21/10/2026
TIME: 14:00
LOCAL: Videoconferência
TITLE:

An Intelligent Tutoring System for Pedagogical Support in the Writing of Game Design Documents with Generative Artificial Intelligence and Prompt Engineering

KEY WORDS:

educational technology; structured writing; pedagogical support; educational games; technology-mediated learning

PAGES: 123
BIG AREA: Ciências Exatas e da Terra
AREA: Ciência da Computação
SUBÁREA: Metodologia e Técnicas da Computação
SPECIALTY: Engenharia de Software
SUMMARY:

This thesis investigates pedagogical support for the writing, revision, and improvement of Game Design Documents (GDDs) in educational contexts, considering difficulties related to the structuring of ideas, the organization and completeness of documents, and the articulation between game design and learning objectives. The objective of this work is to support teachers and students in this structured writing process, fostering the production of GDDs that are clearer, better organized, more coherent, more complete, and aligned with pedagogical objectives. To this end, Vocalis was developed as an Intelligent Tutoring System based on Generative Artificial Intelligence and Prompt Engineering, designed to support content organization, provide formative suggestions, enable successive refinements, and offer evidence for monitoring the development of the documents, while preserving student authorship and teacher mediation. The research is applied in nature, adopts a mixed-methods approach, and follows a hypothetico-deductive orientation. The methodological path comprised a theoretical foundation, a systematic literature review, the development and refinement of the solution, a pilot study in higher education, and a main study conducted as a case study with embedded units in a first-year high school class. The main study was organized into two application cycles involving 39 students distributed across 10 groups in the context of a Chemistry course. The evaluation considers system usage records, GDD versions, refinement prompts, questionnaires administered to participants, and an independent evaluation conducted by a human expert, addressing technical, pedagogical, and usability aspects, the evolution of document quality, and the convergence between automated and human evaluations. The results obtained provide evidence regarding the use of Vocalis as support for the GDD writing and refinement process, enabling the analysis of the evolution of the documents, participants’ perceptions, and the correspondence between the evaluations performed by the system and by a human expert. Overall, the findings indicate possibilities for using Generative Artificial Intelligence as a pedagogical support resource for structured GDD writing, while maintaining student authorship and teacher mediation as central elements of the process.

COMMITTEE MEMBERS:
Presidente - 1671962 - EDUARDO HENRIQUE DA SILVA ARANHA
Interna - 1754430 - LYRENE FERNANDES DA SILVA
Externo à Instituição - FRANCISCO DANTAS DE MEDEIROS NETO - UERN
Externo à Instituição - KLEBER TAVARES FERNANDES - UFERSA
Externo à Instituição - THIAGO REIS DA SILVA - IFMA