NCNewarney Torrezão da Costa

Newarney Torrezão da Costa

*This is an algorithmically generated author record based on publicly available information.

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


Newarney Torrezão da Costa has a total of 93 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    From Design to Classroom Adoption: Teachers' Perspectives on AI‐powered Writing Analytics in Resource‐Constrained Contexts

    by Fabíola Goncalves C. Ribeiro, Luiz Rodrigues, Cleon Xavier, Newarney Torrezão da Costa, Anderson Pinheiro, Jamilla Lobo, Dragan Gasevic, Hendrik Drachsler, Rafael Ferreira Mello

    Background: The adoption of automated tools to support teaching and learning processes has been widely discussed in the literature. However, there remains a need for user‐oriented designs that better align with teachers' practices and classroom realities while providing actionable insights to support instructional choices in authentic educational settings. Objectives: This study aims to systematically evaluate the usability, acceptability and pedagogical relevance of an AIED‐Unplugged learning analytics tool from teachers' perspectives, in order to guide the development of user‐centered design strategies and support writing practices in resource‐constrained educational settings without deepening digital inequities. The proposed study also aims to derive pedagogical and user‐centered design recommendations to strengthen the development of AIED‐U solutions for resource‐constrained educational settings. Method: We employed a mixed‐methods approach involving nine teachers, who downloaded an AIED‐U application and followed a predefined usage scenario according to a think‐aloud protocol. Afterwards, participants filled out the Technology Acceptance Model (TAM) and System Usability Scale (SUS) questionnaires, taking part in a moderated semi‐structured interview guided by an interview script. Results and Conclusion: The study shows how AI‐driven educational tools can support personalization and the development of essay‐writing activities in resource‐constrained contexts while optimizing teachers'time and contributing to students' writing development. The quantitative findings indicated high levels of acceptance and usability, corroborating the qualitative findings that teachers perceived the solution as useful, easy to integrate and supportive of their practices. Taken together, these findings provide empirical evidence of teachers' perceptions regarding the tool's usability, usefulness and potential applicability within authentic classroom workflows in resource‐constrained educational environments.

  • Um Estudo Bibliométrico do Simpósio Brasileiro de Informática na Educação (2001–2025)

    by Cleon Xavier, Patricia Jaques, João Ângelo Nunes, Leonardo Silva, Luiz Rodrigues, Newarney Costa, Rafael Dias Araújo, Sean Siqueira, Rafael Mello

    Este artigo apresenta um estudo bibliométrico sobre o Simpósio Brasileiro de Informática na Educação (SBIE), analisando as publicações entre 2001 e 2025. A pesquisa investigou autores, instituições, regiões e tendências temáticas da comunidade de Informática na Educação no Brasil. A análise identificou forte concentração de publicações nas regiões Sul e Sudeste. Redes de colaboração institucional mostraram comunidades relacionadas à proximidade geográfica. Em relação aos temas, Pensamento Computacional e Ensino de Programação foram alguns tópicos identificados como emergentes. Em contrapartida, temas como Aprendizagem Colaborativa e EaD apresentaram tendência de declínio. Os resultados contribuem para a compreensão da evolução histórica, regional e temática da Informática na Educação no Brasil.

  • Automatic Knowledge Component Extraction to Bridge High School Knowledge Gaps for Engineering Students

    by Mario Antonio Pessoa Santos, Mariana Nassif, Tyagi Lima, Filipe Dwan Pereira, André Nascimento, Luiz Rodrigues, Cleon Xavier, Newarney Torrezão Costa, Gabriel Alves, Rafael Ferreira Mello

    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.

  • Cognitive Scaffolding: An AI-Generated Intelligent Tutoring System Advancing Bloom's Taxonomy in Distance Higher Education

    by Mariana Nassif, Mario Antonio Pessoa Santos, Tyagi Lima, Luiz Rodrigues, Cleon Xavier, Newarney Torrezão Costa, Rafael Ferreira Mello

    In distance learning courses for higher education, bridging the gap between passive reading and active knowledge construction is a critical challenge. This paper presents the design and large-scale deployment of a Cognitive Scaffolding Intelligent Tutoring System (CS-ITS). Unlike generic conversational agents or static signaling tools, CS-ITS operationalizes Bloom's Taxonomy as a control mechanism to orchestrate "upward cognitive scaffolding." The system detects key concepts in digital textbooks, anchors them with re-explanations at the learner's current cognitive level, and automatically generates follow-up questions at the subsequent higher Bloom level. This design transforms static reading into a dynamic inner loop of active learning. We report on a pilot study that involved 7,192 undergraduate students in four disciplines (Pedagogy, Administration, HR, and Logistics). The results indicate that while voluntary participation was 7%, the system achieved high user satisfaction (CSAT 4.7/5) and successfully guided 14% of the engaged users through higher-order cognitive tasks. Critically, the engaged students demonstrated gains in academic performance of up to 2.9%.

  • Open Access

    Evaluating an AI-Powered Learning Analytics Tool for Automated Essay Scoring in Real Classroom Contexts: An Experience Report from a Brazilian Public Secondary School

    by Fabíola G. C. Ribeiro, Luiz Rodrigues, Newarney T. Costa, Cleon Xavier, Jamila Lobo, Anderson P. Cavalcanti, Dragan Gasevic, Hendrik Drachlser, Rafael F. L. Mello

    This experience report, drawn from firsthand classroom practice, presents teachers’ perceptions, opinions, and experiences regarding the adoption of the Escreva Mais, an Artificial Intelligence (AI)-powered Learning Analytics (LA) tool designed for low-resource educational settings and essay assessment. The proposed experience report was conducted in a Brazilian secondary school and involved an essay teacher and a pedagogical coordinator. Both stakeholders adopted the solution through a complete essay assessment and feedback cycle involving thirty-five students. The qualitative findings highlight the Escreva Mais tool pedagogical benefits, perceived usefulness, reliability, performance, and potential contributions to student learning.

  • Feedback Personalizado em Escala no Ensino Superior: Relato de Experiência com a Plataforma Tutoria em Múltiplas Turmas

    by Newarney T. Costa, Cleon Xavier, Luiz Rodrigues, Fabíola G. C. Ribeiro, Taciana Pontual Falcão, Rafael Ferreira Mello

    Fornecer feedback personalizado no ensino superior é um desafio agravado pela sobrecarga docente e turmas numerosas. Este artigo relata a experiência de professores com a plataforma Tutoria, que combina marcação por tags, similaridade semântica e grandes modelos de linguagem para apoiar a correção de questões discursivas, gerar feedback personalizado e prover apoio automatizado às práticas docentes. A plataforma foi utilizada em três turmas, por dois docentes de duas instituições. Também é discutida a percepção discente, coletada por meio de um questionário em quatro dimensões. O relato sugere viabilidade de uso da IA generativa na avaliação formativa para ensino superior em contextos distintos.

  • Open Access

    Which Feedback Is More Effective? A Comparative Study on Teachers' Evaluation of Feedback Models in Essay Writing

    by Anderson Pinheiro Cavalcanti, Luiz Rodrigues, Cleon Xavier, Newarney Torrezão da Costa, Fabiola Ribeiro, Luiz Felipe Bagnhuk Silva, Weslei Chaleghi de Melo, Letícia Santana Stacciarini, Guilherme Weber, Evelyn Vieira, et al.

    Feedback is a key factor in improving the writing skills of students, but providing it at a large scale remains a significant challenge. Large Language Models (LLMs) emerge as a promising solution; however, the pedagogical quality of the automatically generated feedback requires further investigation. This study examined how effective teachers perceived feedback texts generated by an LLM when the prompts were designed to reflect different aspects of feedback theory. To this end, 450 feedback texts were generated for 30 essays and evaluated by five teachers with varying feedback literacy profiles. The results showed that teachers' preferences for feedback models were strongly shaped by their profiles, indicating that there was no single universally "best" feedback model. From a learning analytics perspective, these findings demonstrate how combining theory-informed prompt design with evaluations from diverse teacher profiles can generate actionable insights for the development of scalable, AI-powered feedback systems. The study contributes to learning by showing that the effectiveness of LLM-based feedback depends on personalization not only for students but also for teachers, highlighting new directions for designing adaptive, equitable, and context-sensitive feedback analytics.

  • Large-Scale Implementation of AI-Designed Study Guides and Personalized Feedback in Distance Education

    by Mariana Mello, Mario A. P. Santos, Tyagi Lima, Luiz Rodrigues, Cleon Xavier, Newarney Costa, Rafael Mello

  • Understanding Teacher Revisions of Large Language Model-Generated Feedback

    by Conrad Borchers, Luiz Rodrigues, Newarney Torrezão da Costa, Cleon Xavier, Rafael Ferreira Mello

  • Translating XAI Into Actionable Feedback Using LLMs to Prevent Student Dropout

    by Filipe Dwan Pereira, George Zambonin, André Nascimento, Mario A. P. Santos, Mariana Mello, Tyagi Lima, Luiz Rodrigues, Cleon Xavier, Newarney Torrezão da Costa, Dragan Gašević, et al.