The upkeep of roadside vegetation is essential for ensuring the safety of both motorists and pedestrians. However, identifying the necessity for such maintenance is frequently a time-consuming and costly process, with a high potential for errors in annotation. This work therefore proposes the development of a solution capable of estimating the height vegetation along roadsides in an automated manner. A machine learning model was developed and evaluated using dataset of manually annotated data. The model employs a convolutional neural network architecture, adapted for the task of classifying vegetation heights. The results demonstrate that the model is able to detect the different vegetation height classes with low error rates, indicating its potential for automating the decision-making process for mowing vegetation. This study contributes to the advancement of road monitoring techniques, providing greater operational efficiency and reducing costs for road maintenance.

Machine learning tools are widely used in support of bioacoustics studies, and there are numerous publications on the applicability of convolutional neural networks (CNNs) to the automated presence-absence detection of species. However, the relation between the merit of acoustic background modelling and the recognition performance needs to be better understood. In this study, we investigated the influence of acoustic background substance on the performance of the acoustic detector of the White-lored Spinetail (Synallaxis albilora). Two detector designs were evaluated: the 152-layer ResNet with transfer learning and a purposely created CNN. We experimented with acoustic background representations trained with season-specific (dry, wet, and all-season) data and without explicit modelling to evaluate its influence on the detection performance. The detector permits monitoring of the diel behaviour and breeding time of White-lored Spinetail solely based on the changes in the vocal activity patterns. We report an advantageous performance when background modelling is used, precisely when trained with all-season data. The highest classification accuracy (84.5%) was observed for the purposely created CNN model. Our findings contribute to an improved understanding of the importance of acoustic background modelling, which is essential for increasing the performance of CNN-based species detectors.

A banana desempenha um papel significativo na fruticultura e, para reduzir as perdas durante o processo pós-colheita e prolongar a vida útil de armazenamento, é importante identificar os seus níveis de maturação, principalmente por apresentarem uma alta taxa de deterioração. Técnicas de inteligência artificial podem ser aplicadas para este fim. Este trabalho aborda a construção de modelos de redes neurais artificiais para a classificação de estágios de maturação das bananas, utilizando-se como conjunto de imagens de Bananas Prata Catarina. Foram aplicadas técnicas de data augmentation e redes convolucionais para alcançar melhores resultados.

The public sector has many challenges when improving business processes. One of those challenges is managing the projects' success by considering the context and constraints of public management and business process management (BPM). Moreover, the importance of the elements that will drive success can be different depending on the stakeholder and the stage of the project (e.g. initiation, executing, closing), which adds another ingredient of complexity to project management. Aiming to give project management the capacity to manage success criteria and success factors in BPM projects, this research used action research to develop an integrated model of a project management methodology (PRINCE2) and Success Management. A government-to-government BPM project was conducted as a case study to evaluate the model developed. In addition to the integrated model, another contribution of this work is the proposal of two general lists of success criteria and success factors from the perspectives of the client and project team.

A National Fire Protection Association (NFPA) recomenda que os Bombeiros possuam um valor mínimo de consumo máximo de oxigênio (V̇O2max) de 42 ml/kg/min. Portanto, o objetivo deste foi estimar o V̇O2max de Bombeiros Militares a partir do teste de corrida de 2400m. Participaram do estudo 54 Bombeiros Militares do Estado de Santa Catarina. Foi realizado o teste de corrida de 2400m no menor tempo possível para estimar o V̇O2max. Foi utilizado o coeficiente de correlação de Pearson para verificar a correlação entre as variáveis, adotando-se nível de significância de 5%. A média do V̇O2max estimado foi de 47,6 ± 4,2 ml/kg/min. O tempo médio para realização do teste foi de 658,7 ± 72,4 segundos, com uma velocidade média de 13,3 ± 1,3. Os valores estimados de V̇O2max no presente estudo estão de acordo com os níveis recomendados para essa população. Portanto, o teste de 2400m é capaz de indicar aqueles que apresentam risco elevado de complicações cardíacas, baseado na exigência mínima para o desempenho eficiente e seguro nas atividades de bombeiros, em especial o combate a incêndios. Palavras chaves: V̇O2max, Bombeiros, desempenho.

PurposeAs in the private sector, public organizational information systems (IS) development is commonly carried out through projects. One of the alternatives followed by governmental organizations to perform their projects is outsourcing (by hiring other public institutions that have expertise in the IS area of the projects to be developed). However, limited research has been conducted on project success regarding these government-to-government (G2G) contexts. Since achieving success is crucial for public management, this paper proposes a model for Success Management of IS projects in G2G context.Design/methodology/approachThe research method was design science research (DSR). In the evaluation step of the DSR, IS projects in a G2G environment were the object of case studies.FindingsThis work presents in detail how Success Management activities can be integrated into the processes and process groups of the Project Management Institute's project management guide. The authors also suggest tools and techniques to be used in each Success Management activity.Practical implicationsManaging success, particularly addressing success criteria and success factors, can help managers focus their efforts on what will really impact the success of a project. In the context of IS projects in G2G contexts, this contributes to decreasing waste and increasing the chances of providing better services to citizens.Originality/valueThis work contributes to theory by providing a new model for IS G2G projects that integrates Success Management and project management processes.

A estimativa de localização de objetos no transporte contribui para sistemas de monitoramento de rodovias e até o avanço de veículos autônomos. Este artigo apresenta a criação de um conjunto de dados abrangente e de alta qualidade, destinado ao desenvolvimento e treinamento de modelos de inteligência artificial. Os dados foram coletados, processados e organizados para fornecer informações de coordenadas geográficas do ponto de origem e do objeto alvo, além de informações de posição e tamanho do objeto na imagem. O conjunto de dados possui 9.000 registros, separados entre conjunto de treinamento e de teste, e está disponibilizado publicamente.

To create a bird classification model, it is necessary to have training datasets with thousands of samples. Automating this task is possible, but the first step is being able to segment soundscapes by identifying bird vocalizations. In this study, we address this issue by testing four methods for audio segmentation, the Librosa Library, Few-Shot Learning technique: the BirdNET Framework, and a Bird Classification Model called Perch. The results show that the best method for the purpose of this work was BirdNET, achieving the highest values for precision, accuracy, and F1-score.

É comum o uso de sensoriamento remoto para compreender a dinâmica de desmatamentos. Entretanto, a obtenção e processamento dessas imagens não é uma tarefa trivial, além de haver a dificuldade de ter áreas validadas de desmatamento. Para resolver esse problema este trabalho apresenta o Remote Sensing Deforestation Dataset (RSDD) que contém dados tabulados e imagens de áreas afetadas pelo desmatamento validado por um órgão Estadual. O RSDD possui um tamanho menor que 1% em relação às imagens originais de sensoriamento remoto, proporciona a redução de esforços em futuras pesquisas e pode servir como uma fonte comum para ser utilizada como benchmark em soluções de aprendizado de máquina.

O uso da técnica mais adequada é crucial em qualquer processo Aprendizagem de Máquina. Na bioacústica, devido à complexidade dos dados muitas técnicas tem sido aplicadas e desenvolvidas. Nesse contexto, esse trabalho apresenta a revisão sistemática realizada utilizando a metodologia PRISMA para Bioacústica no monitoramento ambiental por meio de vocalização de pássaros. A revisão demonstrou que as técnicas de Spiking Neural Network, Convolutional Neural Network e Residual Neural Network ganharam destaque nos últimos anos.

Diversos trabalhos foram realizados para o reconhecimento de COVID-19 por meio de imagens de raio-X. Os trabalhos obtinham bom desempenho no reconhecimento das imagens, no entanto, os modelos estão alinhados aos conjuntos de dados utilizados, o que não implica o mesmo desempenho fora do contexto de treino. Deste modo, este trabalho aplicou uma forma justa para avaliar modelos em diferentes cenários. Os resultados demonstraram que os modelos conseguiram distinguir entre diferentes conjuntos de dados de origem diferente, assim, foi determinado que os trabalhos realizados estiveram adaptados ao contexto do conjunto de dados obtido.

Resumo Diante da complexidade dos fenômenos que levam à evasão, foi sistematizado um método que simplifica a elaboração de um diagnóstico, permitindo estruturar e agrupar as dificuldades mais relatadas, possibilitando relacionar as estratégias mais adequadas para contribuir com a permanência e a conclusão de estudantes. A pesquisa caracteriza-se como exploratória e descritiva, utilizando a análise documental para subsidiar a elaboração de questionários estruturados e objetivos. Para aplicação, empregou-se um método não probabilístico, no qual obteveram-se 5.041 respondentes de todos os cursos de graduação da UFMT (33% desvinculados, 37% vinculados e 29% egressos), totalizando cerca de 10% da população entre 2013 e 2019. A ponderação e o agrupamento de dificuldades possibilitou capturar diferentes perspectivas dos discentes. As dificuldades relacionadas com o aprendizado, a estrutura do curso e com o corpo docente foram muito citadas, permitindo a priorização de estratégias para gestores dos cursos e da instituição.

Meteorological elements can affect the environment and cultures differently and may alter the natural development process contributing significantly to climate change. Meteorological variables of the Brazilian Pantanal were studied and used to determine evapotranspiration with fewer variables. It was found that artificial intelligence can substantially improve environmental modeling when alternative prediction techniques are used, resulting in lower project costs and more reliable results. This work tried to find the best combination by comparing machine learning techniques such as artificial neural networks, random forests, and support vector machines. A new model was created that depends on fewer climatic variables compared to the Penman–Monteith method (the standard method for estimating reference evapotranspiration) and can efficiently describe the reference evapotranspiration. Machine learning techniques are highly efficient for modeling environmental systems since they can process large amounts of data and find the best interactions between the parameters involved. In addition, more than 98% accuracy was obtained using fewer variables compared to the standard method when artificial neural networks are utilized.

As perícias criminais em Mato Grosso são realizadas pela Perícia Oficial e Identificação Técnica de Mato Grosso (POLITEC-MT). Este órgão possuía quatro diferentes sistemas para gerenciar as requisições e laudos periciais, trazendo dificuldades de uso, expansão, integração de dados e desempenho. Este trabalho relata a análise desse cenário e o desenvolvimento de um novo sistema, que tem como objetivo facilitar os processos internos da POLITEC-MT por meio de um software com boa usabilidade e eficiência.

Há desigualdade no mercado de trabalho com relação a gênero, o que inclui a área de Tecnologia da Informação. Estudos mostram que a participação feminina é menor e, portanto, ações são necessárias para ajustar esse cenário. Neste trabalho é relatado uma experiência de detecção de um problema envolvendo discentes do sexo feminino em um processo seletivo, e as ações aplicadas para mitigar o respectivo problema, desde um novo processo seletivo até o acompanhamento das novas colaboradoras durante a execução de suas atividades. Espera-se com este trabalho que seja dada atenção a casos semelhantes, além de sugerir ações a serem aplicadas caso o cenário se repita.

Many stages of information systems projects developed in public institutions are defined by laws and regulations, which reduces the management flexibility. In the particular case of Business Process Management (BPM) projects, it has even more influence since the business processes are also constrained by legislation. In this context, it is important to have a clear vision of what the project’s success means for all the stakeholders, what can impact the success, and how success should be evaluated. This paper presents the case of a BPM project of a public institution, where it is being implemented a new PRINCE2-based project management approach which comprises success management activities. The preliminary results include a new model that integrates success management and the PRINCE2 methodology, as well as a set of success criteria and success factors identified for the project.

ABSTRACTThe purpose of this study was to verify the heart rate variability (HRV) and heart rate (HR) kinetics during the fundamental phase in different intensity domains of cycling exercise. Fourteen males performed five exercise sessions: (1) maximal incremental cycling test; (2) two rest‐to‐exercise transitions for each intensity domain, that is, heavy (Δ30) and severe (Δ60) domains. HRV markers (SD1 and SD2) and HR kinetics in the fundamental phase were analyzed by first‐order exponential fitting. There were no significant differences in amplitude values between SD1Δ30 (8.98 ± 3.52 ms) and SD1Δ60 (9.44 ± 3.24 ms) and SD2Δ30 (24.93 ± 9.16 ms) and SD2Δ60 (25.98 ± 7.29 ms). Significant difference was observed between HRΔ30 (52 ± 7 bpm) and HRΔ60 (63 ± 8 bpm). The time constant (τ) values were significantly different between SD1Δ30 (17.61 ± 6.26 s) and SD1Δ60 (13.86 ± 5.90 s), but not between SD2Δ30 (20.06 ± 3.73 s) and SD2Δ60 (19.47 ± 6.03 s) or HRΔ30 (56.75 ± 18.22 s) and HRΔ60 (58.49 ± 15.61 s). However, the τ values for HRΔ30 were higher and significantly different in relation to SD1Δ30 and SD2Δ30, as well as for HRΔ60 in relation to SD1Δ60 and SD2Δ60. The kinetics of the autonomic variable (SD1 marker) was accelerated by the increased intensity. Moreover, significant differences were found for the τ values, with faster HRV markers than HR, in both intensities of Δ30 and Δ60, which suggests that these variables indicate distinct and specific cardiac autonomic response times during different intensity domains in cycling.HIGHLIGHTS The application of HRV to optimize exercise prescription at different effort intensities is extremely important to obtain assertive and effective results. Analysis of the kinetic responses of HRV is a useful tool for the evaluation of exercise performance and health status. A faster kinetics was found for HRV markers in comparison to HR, for both intensities analysed, which suggests that these variables indicate distinct and specific cardiac autonomic response times during different intensity domains in cycling.

Automated acoustic recognition of birds is considered an important technology in support of biodiversity monitoring and biodiversity conservation activities. These activities require processing large amounts of soundscape recordings. Typically, recordings are transformed to a number of acoustic features, and a machine learning method is used to build models and recognize the sound events of interest. The main problem is the scalability of data processing, either for developing models or for processing recordings made over long time periods. In those cases, the processing time and resources required might become prohibitive for the average user. To address this problem, we evaluated the applicability of three data reduction methods. These methods were applied to a series of acoustic feature vectors as an additional postprocessing step, which aims to reduce the computational demand during training. The experimental results obtained using Mel-frequency cepstral coefficients (MFCCs) and hidden Markov models (HMMs) support the finding that a reduction in training data by a factor of 10 does not significantly affect the recognition performance.