Efficient public transportation management is essential for the development of large urban centers, providing several benefits such as comprehensive coverage of population mobility, reduction of transport costs, better control of traffic congestion, and significant reduction of environmental impact limiting gas emissions and pollution. Realizing these benefits requires a deeply understanding the population and transit patterns and the adoption of approaches to model multiple relations and characteristics efficiently. This work addresses these challenges by providing a novel dataset that includes various public transportation components from three different systems: regular buses, subway, and BRT (Bus Rapid Transit). Our dataset comprises daily information from about 700,000 passengers in Salvador, one of Brazil’s largest cities, and local public transportation data with approximately 2,000 vehicles operating across nearly 400 lines, connecting almost 3,000 stops and stations. With data collected from March 2024 to March 2025 at a frequency lower than one minute, SUNT stands as one of the largest, most comprehensive, and openly available urban datasets in the literature.

Abstract Video surveillance has benefited greatly from advances in artificial intelligence, particularly in computer vision, and the number of monitored environments has consequently increased, creating new challenges in extracting relevant information. When multiple cameras cover adjacent areas, overlapping fields of view can cause the same subject to be detected across cameras, introducing duplicate counts. Although the literature offers established solutions, many rely on high-quality recordings and complex, computationally intensive methods, limiting their use on resource-constrained devices. In this work, we address these challenges with a solution that leverages minimal information about target subjects and uses a transductive strategy to detect duplicates based on interactions within overlapping fields of view. Experiments in real-world settings show that our approach suppresses duplicates effectively, making it suitable for deployment on resource-limited hardware. We evaluate state-of-the-art lightweight models with high inference speed, as well as classical re-identification methods, in scenarios with low-quality video and constrained devices. The results underscore the effectiveness of the proposed approach and motivate exploration of re-ID with domain adaptation, as well as anchor-free methods with weak or semi-supervised learning.

The widespread adoption of Machine Learning (ML) is leading to an increase in processing demands. Dealing with the growing volume of data poses a significant challenge in providing accurate classification services using ML models. Offloading ML tasks to network switches presents an opportunity to tackle this challenge, offering high throughput and low latency. Nonetheless, network devices encounter limitations in resources, and programmable languages like P4 lack support for basic operations, necessary for the ML methods, including floating-point arithmetic and native repetition structures. In this paper, we investigate the use of drift detection ML models to enhance the accuracy of in-network traffic classification. The novelty lies in designing drift detection based on bitwise operations, which are well-suited for implementation within the data plane. As a proof-of-concept, we implement drift detection using the P4 language on BMv2 switches, validated with a dataset of over 2 million samples. Our results demonstrate a significant increase in classification accuracy with drift detection, while maintaining line-speed operation and quickly adapting to changes in traffic patterns.

Hemophilia is a genetic disorder characterized by mutations in coagulation factor proteins, resulting in impaired blood coagulation. Despite recent advances in diagnosing and treating Hemophilia, the precise function of each residue within such proteins remains unclear. In this work, aiming to provide a crucial step to addressing this complex problem, we present a systematic approach that organizes the coagulation factor proteins into graphs to be later modeled by novel Graph Neural Networks (GNN). In our graphs, nodes represent amino acids, which can be connected by proximity in the three-dimensional structure. Our experiments on FVIII (coagulation factor VIII) proteins, responsible for causing Hemophilia A, presented significant results in predicting severe and mild forms of the disease. Moreover, focused on supporting the advance of novel recombinant therapeutic FVIII proteins, we adapted our study to predict the activity and expression of more than 300 in vitro alanine mutations. Besides innovating the form in which hemophilia is organized and modeled, this work presents a GNN architecture specifically designed to optimize the classification of FVIII proteins. Our results emphasize how graph-based approaches support a deeper understanding of Hemophilia A, thus providing a solid scientific basis for further research in this field.

The Intergovernmental Panel on Climate Change and the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services have emphasised unequivocal evidences about the impact of human actions on climate and biodiversity at alarming rates. In Brazilian terms, 2019 has been marked by controversial discussions among politicians and environmentalists, leading to misinformation and misinterpretations that clearly motivate the continuous collection and scientific analysis of data to support sustainable solutions. Aiming at dealing with this issue, this manuscript brings two contributions: (i) the creation of the Brazilian Forest Dataset, including Brazilian seed plants, Fraction of Absorbed Photosynthetically Active Radiation, meteorological and geographical data composing 8,482 attributes to model and predict 20 vegetation types; and (ii) the feasibility analysis on modelling this dataset in light of supervised machine learning algorithms, so we devise confident results on the Brazilian biodiversity. Experimental results confirm Random Forest and Support Vector Machines successfully adjust models, enabling researchers to predict the occurrence of specific types of vegetation in different regions of Brazil as well as analyse how the prediction accuracy changes along time after the collection of new data. Our contributions bring important tools to support the study on the evolution of the Brazilian biodiversity.

Source code quality plays a key role in software quality mainly due to its impact on software maintainability. Software engineers have been using source code metrics to support them to assess source code quality. Source code metrics quantify different source code characteristics. However, source code metric analysis still involves subjectivity. For instance, it is not trivial to decide whether a metric value is high or low. To reduce the eventual subjectivity of source code metrics analysis, several researchers are using Machine Learning algorithms. Therefore, in this paper, we designed a Fuzzy-based approach to extract characteristics and patterns present in source code versioning repositories in order to: i) assist the specialist in the interpretation of releases, especially when working with large volumes of source code; ii) from the release interpretation, specialists can improve the quality of the source code; and iii) monitor the evolution of the software as new releases are submitted to the repositories. We evaluated the proposed approach with the Linux Test Project repository, emphasizing the interpretability of large source code versioning repositories.

In this manuscript, we performed a scrutiny on hepatitis records collected in Brazil using statistical and computational tools aiming at analyzing the evolution of this disease during the period of 2007 and 2014. The study was performed in three phases. Initially, we used statistical tests to assess whether or not the notification cases were increasing. Then, we created a measure to quantify trends, without being affected by scale, thus allowing to compare the occurrence of hepatitis in different regions in Brazil. Finally, we used concept drift to detect the changing point when the notification behavior has changed over time. As a consequence of this study, public policies can be developed to deal with the increasing number of cases, specially by considering sexual transmissions.

The monitoring of volcanoes is mainly performed by sensors installed on their structures, aiming at recording seismic activities and reporting them to observatories to be later analyzed by specialists. However, due to the high volume of data continuously collected, the use of automatic techniques is an important requirement to support real time analyses. In this sense, a basic but challenging task is the classification of seismic activities to identify signals yielded by different sources as, for instance, the movement of magmatic fluids. Although there exists several approaches proposed to perform such task, they were mainly designed to deal with raw signals. In this paper, we present a 2D approach developed considering two main steps. Firstly, spectrograms for every collected signal are calculated by using Fourier Transform. Secondly, we set a deep neural network to discriminate seismic activities by analyzing the spectrogram shapes. As a consequence, our classifier provided outstanding results with accuracy rates greater than 95%.