Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Waldemiro José Assis Gomes Negreiros has a total of 45 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    EcoLex RAG: Sistema HyPA-RAG com Legal-BERTimbau para Suporte Explicável à Decisão em Gestão Ambiental Brasileira

    by Waldemiro Negreiros, Delvek Sousa, Leonardo Estevam, Suellen Negreiros, Vitor Melo, Filipe Corrêa, Saulo Oliveira, Carlos Motta, Paulo Pontes, Marcos Seruffo

    A governança ambiental pós-COP30 exige agilidade na consulta ao complexo arcabouço legislativo brasileiro. Este artigo apresenta o EcoLex RAG, um sistema HyPA-RAG triplo (BM25, Legal-BERTimbau e Knowledge Graph) focado no suporte explicável à decisão. Controlado por um classificador adaptativo de complexidade, o sistema alcança 65% de Law Retrieval Rate (LRR) e 48% de Citation Article Rate (CAR). A abordagem reduz o consumo de contexto em 32% (de 10,0 para 6,8 fontes em média), superando métodos estáticos. O sistema viabiliza uma inteligência artificial eficiente, estritamente auditável e de alto impacto estratégico para a administração pública.

  • Open Access

    Fusion of Orbital and Surface Data for Air Temperature Estimation in Belém Using Machine Learning

    by Leonardo de O. Tamasauskas, Waldemiro J. A. G. Negreiros, Pedro H. Do V. Guimarães, Williane G. S. Pereira, Jean A. C. Dias, Gabriel B. Costa, Marcos C. da R. Seruffo

    This study evaluates fusion strategies between ERA5-Land orbital data and INMET surface observations to estimate daily air temperature in Belém, using data from 1994 to 2024. Different Random Forest models were compared, along with SARIMAX, LSTM, and stacking (SARIMAX and RF). The RF model with multivariate fusion achieved the best performance (RMSE = 0.5157 °C; R2 = 0.708; Skill = 6.19%), reducing bias relative to the orbital baseline, while chronological validation confirmed the temporal stability of RF-based models. The results indicate that, under low equatorial thermal variability, tree-based models are the most robust for satellite–surface fusion and thermal downscaling.

  • Open Access

    Machine Learning-Based Spatial Modeling of Land Surface Temperature in Butantã: Insights from SHAP Interpretability

    by Williane G. S. Pereira, Gabriel V. de Melo, Waldemiro J. A. G. Negreiros, Fernando A. R. Costa, Leonardo De O. Tamasauskas, Claudomiro De S. de S. Junior, Marcos C. R. Seruffo

    Accelerated urbanization has expanded urban heat islands, necessitating the precise monitoring of Land Surface Temperature (LST). This study spatially predicts LST in Butantã (SP) for 2018 and 2021 using a high-resolution multiscale approach. To capture the relationships between urban morphology and LST, XGBoost and LightGBM models were compared, with XGBoost demonstrating superior predictive performance (R2 of 0.890 and 0.848). Applying the SHapley Additive exPlanations (SHAP) method to the winning model ensured explainability, revealing the dominance of impervious surfaces in warming, the strong mitigating potential of water bodies and vegetation, and the heat retention induced by topography in valleys.

  • Open Access

    Abordagem Multivariada para Imputação de Temperatura Média na Região Amazônica baseada em Estações Meteorológicas e Dados de Reanálise ERA5-Land

    by Jean Arthur Costa Dias, André Vinicius N. Alves, Waldemiro J. A. G. Negreiros, Rodrigo Gonçalves Novais, Fernando Luiz Cyrino Oliveira, Leonardo De O. Tamasauskas, Pedro H. Do V. Guimarães, Karla Figueiredo, Gabriel B. Costa, Marivan S. Gomes, et al.

    This study investigates the imputation of missing values in mean air temperature time series in the state of Pará, Brazil, by integrating observational data from INMET and the ERA5-Land reanalysis. A multivariate non-linear approach based on the MissForest algorithm was developed and compared with models using only station data and with linear interpolation. The evaluation employed synthetic gaps together with the RMSE, MAE, and MAE in quantile 99 metrics. The results indicate that the inclusion of ERA5-Land reduced RMSE to 0.664–0.744 °C, MAE to 0.509–0.569 °C, and ∆P99 to 0.909–1.013 °C, while increasing model stability and outperforming the other strategies, thus supporting the methodological consistency of the approach for tropical climate time series.

  • Open Access

    Innovations in Geometry Teaching with MiritiBoard VR 2.0 Glasses: Customizations and Usability Evaluation

    by João Francisco Moreira de Carvalho, Flávio Rafael Trindade Moura, Lucas Gabriel Do Carmo Pereira, Leonardo da Conceição Estevam, Waldemiro José Assis Gomes Negreiros, André Vinícius Neves Alves, Lyanh Vinicios Lopes Pinto, André Jordan Santos dos Santos, Walter dos Santos Oliveira Júnior, Diego Lisboa Cardoso, et al.

    The use of Virtual Reality (VR) headsets has rapidly expanded into various areas, going beyond entertainment and electronic games. These devices provide immersive experiences that simulate three-dimensional environments, allowing for greater user interaction and engagement. Given the growth of mobile VR, this study presents MiritiBoard VR, a low-cost headset made from miriti, a plant native to the Amazon, focusing on sustainability, promotion of the local bioeconomy, and geometry education for basic education students in Brazil. The device was used in the educational application GeoMeta, enabling the visualization of 360-degree content, Augmented Reality, and gamification for geometry teaching. In order to improve the user experience of the MiritiBoard VR, two prototypes with controllers were developed to offer more features and comfort to the glasses, as it has limitations in mobility and interactivity. One prototype has a control integrated into the viewer and the other uses an external GamePad, both connected via Bluetooth, which is housed inside the headset. The evaluation, conducted using the System Usability Scale (SUS), showed that although the conventional model achieved the highest overall score, the MiritiBoard with integrated control was the users' preferred choice. The results indicate that MiritiBoard VR offers an immersive experience comparable to more expensive devices, with potential for expansion through new sensors. In addition to contributing significantly to mathematics education, the project reinforces responsible consumption practices and supports the local Amazonian economy.


  • Open Access

    Development and Validation of a Low-Cost Arduino-Based Lee Disc System for Thermal Conductivity Analysis of Sustainable Roofing Materials

    by Waldemiro José Assis Gomes Negreiros, Jean da Silva Rodrigues, Maurício Maia Ribeiro, Douglas Santos Silva, Raí Felipe Pereira Junio, Marcos Cesar da Rocha Seruffo, Sergio Neves Monteiro, Alessandro de Castro Corrêa

    The optimization of thermal performance in buildings is essential for sustainable urban development, yet the high cost and complexity of traditional thermal conductivity measurement methods limit broader research and educational applications. This study developed and validated a low-cost, replicable prototype that determines the thermal conductivity of roof tiles and composites using the Lee Disc method automated with Arduino-based acquisition. Standardized samples of ceramic, fiber–cement, galvanized steel, and steel coated with a castor oil-based polyurethane composite reinforced with miriti fiber (Mauritia flexuosa) were analyzed. The experimental setup incorporated integrated digital thermocouples and strict thermal insulation procedures to ensure measurement precision and reproducibility. Results showed that applying the biocompatible composite layer to metal tiles reduced thermal conductivity by up to 53%, reaching values as low as 0.2004 W·m−1·K−1—well below those of ceramic (0.4290 W·m−1·K−1) and fiber–cement (0.3095 W·m−1·K−1) tiles. The system demonstrated high accuracy (coefficient of variation < 5%) and operational stability across all replicates. These findings confirm the feasibility of open-source, low-cost instrumentation for advanced thermal characterization of building materials. The approach expands access to experimental research, promotes sustainable insulation technologies, and offers practical applications for both scientific studies and engineering education in resource-limited environments.


  • Open Access

    Cloud Segmentation in Multispectral Images from the Sentinel-2 Satellite: A Comparative Study of Deep Learning Approaches

    by Jean A. C. Dias, Williane G. S. Pereira, Waldemiro J. A. G. Negreiros, Pedro H. Do V. Guimarães, Alan B. S. Corrêa, Leonardo de O. Tamasauskas, Marivan S. Gomes, Danilo Souza, William Yoshida, Danilo Valente, et al.

    To enhance the processing of multispectral images under adverse conditions, such as the presence of clouds, this study compared the neural networks U-Net, DeepLabV3+, and SegFormer using images from the Sentinel-2 satellite, evaluating their performance through metrics such as Intersection over Union (IoU) and Dice coefficient. The results indicate that SegFormer achieved the highest accuracy in cloud detection but with a longer inference time, while U-Net demonstrated a balance between accuracy and efficiency, and DeepLabV3+ stood out for its shorter processing time but lower performance. The study highlights the relevance of neural networks in remote sensing and indicates the need for model optimization.


  • Open Access

    Potential Use of Castor Oil-Based Polyurethane Matrix Composite with Miriti Fiber Filling as Thermal Insulation Applied to Metal Tiles

    by Waldemiro José Assis Gomes Negreiros, Jean da Silva Rodrigues, Maurício Maia Ribeiro, Douglas Santos Silva, Raí Felipe Pereira Junio, Sergio Neves Monteiro, Alessandro de Castro Corrêa

    The Amazon biome’s climate, with annual temperatures above 30 °C and humidity over 90%, poses challenges for building thermally comfortable structures without expensive cooling systems. This study developed a castor oil-based polyurethane (PU) composite with miriti fiber (Mauritia flexuosa) as a roof thermal blanket, comparing its performance to fiber cement, ceramic, and metal tiles. Measurements were conducted over 136 days at the Federal Institute of Education, Science and Technology of Pará, Campus Belém. From August to October 2022, the fiber cement tile (CT) showed average thermal reductions of 5.9475 °C, 6.13388 °C, and 6.37368 °C, while the FCT coating had more modest reductions of 3.6634 °C, 3.63291 °C, and 3.60598 °C. In November and December 2023, the PU/miriti coating reached the highest reductions, 18.64058 °C and 17.88021 °C. Meanwhile, FCT recorded lower values of 1.74124 °C and 1.74721 °C. Observations show fiber cement allowed the highest heat transfer, whereas a metal tile combined with the PU/miriti composite provided better thermal performance than fiber cement and ceramic, meeting standards approval. The findings highlight the PU/miriti composite’s viability for roofing in hot, humid climates where maintaining lower indoor temperatures is essential. By reducing reliance on mechanical cooling, this technology can foster sustainable, cost-effective building practices in the region.