Filipe Dwan Pereira
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Filipe Dwan Pereira has a total of 149 co-authors. Here are the top co-authors by number of shared publications:
- Elaine Harada Teixeira de Oliveira
Universidade Federal do Amazonas (UFAM)
32 shared publications - Leandro Silva Galvão de Carvalho
Universidade Federal do Amazonas (UFAM)
24 shared publications - Alexandra Ioana Cristea
Durham University
15 shared publications - Rafael Ferreira Mello
Universidade Federal Rural de Pernambuco (UFRPE)
12 shared publications David Braga Fernandes de Oliveira
UFAM
11 shared publications- Luiz Rodrigues
Universidade Tecnológica Federal do Paraná
11 shared publications
Latest Publications
- 5 October 2026
- Published by Sociedade Brasileira de Computação
- p. 2428-2442
Padrões Semanais de Engajamento em AVA e Evasão em Disciplinas EaD: uma Análise Longitudinal com Big Data Educacional
A evasão na Educação a Distância é um desafio complexo, e seus sinais de descontinuidade podem emergir via padrões semanais de engajamento em Ambientes Virtuais de Aprendizagem (AVA). Este estudo realiza uma análise longitudinal com Big Data Educacional a partir de logs de 39.372 estudantes de graduação, totalizando 224.309 observações semanais. Por meio de Equações de Estimação Generalizadas (GEE), extraíram-se indicadores interpretáveis capazes de modelar o engajamento e o risco de evasão em uma disciplina ofertada em 31 cursos. Os resultados evidenciam padrões comportamentais associados à evasão; destaca-se, por exemplo, que calouros apresentam chance 2,11 vezes maior de evasão da disciplina quando comparados aos veteranos.
- 5 October 2026
- Published by Sociedade Brasileira de Computação
- p. 2428-2442
Padrões Semanais de Engajamento em AVA e Evasão em Disciplinas EaD: uma Análise Longitudinal com Big Data Educacional
A evasão na Educação a Distância é um desafio complexo, e seus sinais de descontinuidade podem emergir via padrões semanais de engajamento em Ambientes Virtuais de Aprendizagem (AVA). Este estudo realiza uma análise longitudinal com Big Data Educacional a partir de logs de 39.372 estudantes de graduação, totalizando 224.309 observações semanais. Por meio de Equações de Estimação Generalizadas (GEE), extraíram-se indicadores interpretáveis capazes de modelar o engajamento e o risco de evasão em uma disciplina ofertada em 31 cursos. Os resultados evidenciam padrões comportamentais associados à evasão; destaca-se, por exemplo, que calouros apresentam chance 2,11 vezes maior de evasão da disciplina quando comparados aos veteranos.
- 5 October 2026
- Published by Sociedade Brasileira de Computação
- p. 716-729
An AI Agent for Democratizing Educational Policy Data
Equitable educational public policies require accessible data, yet much remains locked in unstructured PDFs. This paper presents a conversational AI agent designed to democratize access to a 15,000-page educational policy corpus, prioritizing source traceability, semantic preservation of tabular data, and epistemic humility. Evaluations demonstrate reliability for public administration: a 91.3% expert accuracy rate (63/69 successful queries) and RAGAS baseline metrics for answer relevancy and context precision. Furthermore, the system fosters trust by explicitly abstaining when data is absent and proves robust against prompt injection and data poisoning attacks. These suggest its potential as an active analytical tool to support evidence-based policymaking.
- 5 October 2026
- Published by Sociedade Brasileira de Computação
- p. 716-729
An AI Agent for Democratizing Educational Policy Data
Equitable educational public policies require accessible data, yet much remains locked in unstructured PDFs. This paper presents a conversational AI agent designed to democratize access to a 15,000-page educational policy corpus, prioritizing source traceability, semantic preservation of tabular data, and epistemic humility. Evaluations demonstrate reliability for public administration: a 91.3% expert accuracy rate (63/69 successful queries) and RAGAS baseline metrics for answer relevancy and context precision. Furthermore, the system fosters trust by explicitly abstaining when data is absent and proves robust against prompt injection and data poisoning attacks. These suggest its potential as an active analytical tool to support evidence-based policymaking.
- 5 October 2026
- Published by Sociedade Brasileira de Computação
- p. 2565-2573
Automatic Knowledge Component Extraction to Bridge High School Knowledge Gaps for Engineering Students
Many students entering higher education have knowledge gaps from secondary education that contribute to low retention and high dropout rates. This short paper presents a case study that applies Knowledge Component (KC) extraction and recommendation techniques to help first-year engineering students bridge these gaps. Using embedding-based similarity methods, we mapped K–12 mathematical competencies to higher education prerequisites and delivered personalized preparatory resources through short video lectures and AI-generated summaries. Results indicate that the similarity-based approach achieved high validation rates in STEM-oriented disciplines, reliably identifying prerequisite KCs, while performance was less consistent in non-technical courses.
- 5 October 2026
- Published by Sociedade Brasileira de Computação
- p. 2565-2573
Automatic Knowledge Component Extraction to Bridge High School Knowledge Gaps for Engineering Students
Many students entering higher education have knowledge gaps from secondary education that contribute to low retention and high dropout rates. This short paper presents a case study that applies Knowledge Component (KC) extraction and recommendation techniques to help first-year engineering students bridge these gaps. Using embedding-based similarity methods, we mapped K–12 mathematical competencies to higher education prerequisites and delivered personalized preparatory resources through short video lectures and AI-generated summaries. Results indicate that the similarity-based approach achieved high validation rates in STEM-oriented disciplines, reliably identifying prerequisite KCs, while performance was less consistent in non-technical courses.
- 25 June 2026
- Published by Springer Nature
- p. 582-591
Translating XAI Into Actionable Feedback Using LLMs to Prevent Student Dropout
- 25 June 2026
- Published by Springer Nature
- p. 582-591
Translating XAI Into Actionable Feedback Using LLMs to Prevent Student Dropout
- Top Cited PapersOpen Access
- 3 March 2025
- Published by Association for Computing Machinery (ACM)
- p. 93-103
Automatic Short Answer Grading in the LLM Era: Does GPT-4 with Prompt Engineering beat Traditional Models?
- 3 March 2025
- Published by Association for Computing Machinery (ACM)
- p. 93-103
Top Cited PapersOpen AccessAutomatic Short Answer Grading in the LLM Era: Does GPT-4 with Prompt Engineering beat Traditional Models?
- Open Access
- 4 November 2024
- Published by Sociedade Brasileira de Computação
- p. 1730-1743
Prompt Engineering for Automatic Short Answer Grading in Brazilian Portuguese
Automatic Short Answer Grading (ASAG) is a prominent area of Artificial Intelligence in Education (AIED). Despite much research, developing ASAG systems is challenging, even when focused on a single subject, mostly due to the variability in length and content of students' answers. While recent research has explored Large Language Models (LLMs) to enhance the efficiency of ASAG, the LLM performance is highly dependent on the prompt design. In that context, prompt engineering plays a crucial role. However, to the best of our knowledge, no research has systematically investigated prompt engineering in ASAG. Thus, this study compares over 128 prompt combinations for a Portuguese dataset based on GPT-3.5-Turbo and GPT-4-Turbo. Our findings indicate the crucial role of specific prompt components in improving GPT results and shows that GPT-4 consistently outperformed GPT-3.5 in this domain. These insights guide prompt design for ASAG in the context of Brazilian Portuguese. Therefore, we recommend students, educators, and developers leverage these findings to optimize prompt design and benefit from the advancements offered by state-of-the-art LLMs whenever possible.
- 4 November 2024
- Published by Sociedade Brasileira de Computação
- p. 1730-1743
Open AccessPrompt Engineering for Automatic Short Answer Grading in Brazilian Portuguese
Automatic Short Answer Grading (ASAG) is a prominent area of Artificial Intelligence in Education (AIED). Despite much research, developing ASAG systems is challenging, even when focused on a single subject, mostly due to the variability in length and content of students' answers. While recent research has explored Large Language Models (LLMs) to enhance the efficiency of ASAG, the LLM performance is highly dependent on the prompt design. In that context, prompt engineering plays a crucial role. However, to the best of our knowledge, no research has systematically investigated prompt engineering in ASAG. Thus, this study compares over 128 prompt combinations for a Portuguese dataset based on GPT-3.5-Turbo and GPT-4-Turbo. Our findings indicate the crucial role of specific prompt components in improving GPT results and shows that GPT-4 consistently outperformed GPT-3.5 in this domain. These insights guide prompt design for ASAG in the context of Brazilian Portuguese. Therefore, we recommend students, educators, and developers leverage these findings to optimize prompt design and benefit from the advancements offered by state-of-the-art LLMs whenever possible.
- Open Access
- 8 October 2024
- Published by Sociedade Brasileira de Computação in Revista Brasileira de Informática na Educação
- Vol. 32, 510-532
Brazilian Teachers’ concerns towards the use of Gamification in Education: perceived barriers to its adoption
Gamification applied to learning environments is widely accepted as positively impacting students' psychological and cognitive aspects, such as motivation and learning performance. According to the literature on the subject, gamification tends to promote more positive effects on students than negative ones. Meanwhile, the literature lacks a deeper understanding of how education professionals perceive gamification in learning environments and their concerns about implicit issues and ethical issues. Prior research has not examined the relationship between gamification in education, its ethical concerns, and barriers. As a result, we expanded a previous study to identify and delve deep into potential barriers and ethical concerns pertaining to gamification from the perspective of Brazilian teachers. A survey was designed and answered by 61 Brazilian teachers. According to our findings, teachers are not inclined to use gamification for various reasons, such as social factors (e.g., acceptance by teachers and students) and planning and evaluation issues (e.g., lack of knowledge). Our study also found that their ethical concerns pertain to psychological effects, social issues, privacy issues, humanization, and behavioral effects. As part of the contribution of this paper, we list potential barriers and ethical concerns that designers and researchers should keep in mind when designing and implementing gamification and gamification-based personalization in learning environments.
- 8 October 2024
- Published by Sociedade Brasileira de Computação in Revista Brasileira de Informática na Educação
- Vol. 32, 510-532
Open AccessBrazilian Teachers’ concerns towards the use of Gamification in Education: perceived barriers to its adoption
Gamification applied to learning environments is widely accepted as positively impacting students' psychological and cognitive aspects, such as motivation and learning performance. According to the literature on the subject, gamification tends to promote more positive effects on students than negative ones. Meanwhile, the literature lacks a deeper understanding of how education professionals perceive gamification in learning environments and their concerns about implicit issues and ethical issues. Prior research has not examined the relationship between gamification in education, its ethical concerns, and barriers. As a result, we expanded a previous study to identify and delve deep into potential barriers and ethical concerns pertaining to gamification from the perspective of Brazilian teachers. A survey was designed and answered by 61 Brazilian teachers. According to our findings, teachers are not inclined to use gamification for various reasons, such as social factors (e.g., acceptance by teachers and students) and planning and evaluation issues (e.g., lack of knowledge). Our study also found that their ethical concerns pertain to psychological effects, social issues, privacy issues, humanization, and behavioral effects. As part of the contribution of this paper, we list potential barriers and ethical concerns that designers and researchers should keep in mind when designing and implementing gamification and gamification-based personalization in learning environments.
- 13 September 2024
- Published by Springer Nature
- p. 359-374
From Sparse to Smart: Leveraging AI for Effective Online Judge Problem Classification in Programming Education
- 13 September 2024
- Published by Springer Nature
- p. 359-374
From Sparse to Smart: Leveraging AI for Effective Online Judge Problem Classification in Programming Education
- 2 July 2024
- Published by Springer Nature
- p. 418-426
Enhancing Algorithmic Fairness in Student Performance Prediction Through Unbiased and Equitable Machine Learning Models
- 2 July 2024
- Published by Springer Nature
- p. 418-426
Enhancing Algorithmic Fairness in Student Performance Prediction Through Unbiased and Equitable Machine Learning Models
- 2 July 2024
- Published by Springer Nature
- p. 192-205
Can GPT4 Answer Educational Tests? Empirical Analysis of Answer Quality Based on Question Complexity and Difficulty
- 2 July 2024
- Published by Springer Nature
- p. 192-205
Can GPT4 Answer Educational Tests? Empirical Analysis of Answer Quality Based on Question Complexity and Difficulty