Eulanda Miranda dos Santos

A Floresta Amazônica abriga elevada diversidade vegetal, cuja identificação precisa é fundamental para estudos em taxonomia, ecologia e conservação. Este estudo descreve o desenvolvimento e a avaliação do ES-PLANTUFAM, um sistema especialista online para identificação de famílias de angiospermas da Amazônia brasileira. A base de conhecimento foi construída a partir de literatura taxonômica especializada, e diferentes algoritmos de árvores de decisão foram testados no software Weka para geração das regras do sistema, posteriormente implementadas no ES-Builder. A validação envolveu testes com amostras depositadas no Herbário HUAM/UFAM e avaliação de usabilidade por usuários. O modelo RandomTree apresentou o melhor desempenho (91,6% de acertos; Kappa = 0,91), e os testes com amostras indicaram precisão de 88,89%. O ES-PLANTUFAM demonstrou potencial como ferramenta digital para apoio à identificação botânica, tornando mais acessíveis os caracteres morfológicos diagnósticos utilizados na distinção entre famílias de angiospermas.

Camera trap is an important non-invasive technique for wildlife monitoring. A typical camera-trap workflow involves various relevant tasks, such as filtering empty images, classifying animal species and identifying animal behavior. In this study, we explore the application of large-scale multimodal language models (MLLMs) for processing camera trap images to perform these three tasks. We evaluate the performance of four state-of-the-art models across these tasks, precisely BLIP, CLIP, Gemini, and GPT with zero-shot and few-shot learning methodologies. Our experiments showed several interesting results. First, few-shot learning significantly enhanced model performance in filtering empty images, with BLIP achieving a much higher accuracy (91.0%) compared to only 7.61% of its zero-shot counterpart. In the task of animal species classification, Gemini showed strong baseline performance, reaching 75.89 % of accuracy with zero-shot. In terms of identifying animal behavior, two scenarios were investigated: using single image or sequences of images. The results indicate that sequence-based processing improves behavioral analysis, with BLIP attaining the highest accuracy (75.57 %) in this task. In general, our study emphasizes the limitations of the zero-shot approach in complex tasks while highlights the effective potential of few-shot and sequence-based learning to address challenging problems such as empty images, and species misclassifications. These findings demonstrate the efficacy of advanced MLLMs in automating biodiversity monitoring, offering a scalable and accurate solution for processing large-scale datasets, and advancing conservation science.

Computer vision applications often involve processing large-scale multidimensional data, requiring methods that are both efficient and accurate. Traditional pattern recognition methods based on subspace representations offer low computational complexity but typically underperform compared to deep learning models in terms of recognition accuracy. This study aims to explore and analyze the integration of subspace representations within deep learning frameworks to leverage the advantages of both approaches. We conducted a comprehensive survey of existing methods that combine subspace representation techniques with deep neural networks. We propose a taxonomy to categorize these methods into three distinct groups based on their integration strategies. The reviewed methods demonstrate that incorporating subspace representations can enhance the performance and efficiency of deep learning models. The taxonomy helps to clarify the landscape of these hybrid approaches and identifies trends in methodological development. The surveyed approaches demonstrate a clear methodological evolution, contributing to enhanced outcomes in various real-world applications.

The retrieval of similar soundscapes is essential for bioacoustic and ecoacoustic monitoring, yet it remains challenging due to the large volume of unlabeled data, environmental noise, and the complexity of acoustic scenes. To overcome the limitations of traditional, feature-based methods, this study proposes an efficient system that integrates embeddings extracted from a pretrained deep learning model, combined with a noise reduction technique and feature fusion strategies within a vector database to enable similarity-based retrieval. We evaluated the system using bird, amphibian, and mammal recordings across four experimental methodologies, including a use case focused on endangered species. Results show that embedding vectors consistently outperform traditional MFCC (Mel-frequency cepstral coefficients) features in capturing acoustic similarity, and that approximate search algorithms (HNSW) significantly improve both retrieval precision and query efficiency. Additionally, the system effectively retrieves recordings of the critically endangered species Crax alberti and maps their geographic distribution, highlighting its potential for conservation planning and early-warning monitoring.

A governança digital é um impulsionador crucial da sustentabilidade, ao possibilitar transparência, responsabilidade e eficiência na administração de recursos. Dentre os diversos mecanismos pelos quais a governança digital contribui para o desenvolvimento sustentável, destaca-se seu efeito mediador na política ambiental, pois fornece base para uma gestão sustentável do consumo, alocação de subsídios e proteção de recursos naturais. O foco deste artigo é destacar que sistemas de monitoramento ambiental apoiados por tecnologia digital fortalecem o papel fundamental da governança digital na sustentabilidade ambiental, por favorecerem decisões mais científicas e orientadas por dados.

A summary of the principles and actions that have emerged from the Brazilian Computing Society's global climate change task force, focusing on the importance of computing in addressing sustainability challenges.

Concept drift, a phenomenon that can lead to degradation of classifier performance over time, is commonly addressed in the literature through detection and reaction strategies. However, these strategies often rely on complete classifier retraining without considering the properties of the drift, which can prove inadequate in many scenarios. Limited attention has been given to understanding the nature of drift and its characterization, which are crucial for designing effective reaction strategies. Drift descriptors provide a means to explain how new concepts replace existing ones, offering valuable insights into the nature of drift. In this context, this work examines the impact of four descriptors—severity, recurrence, frequency, and speed—on concept drift through extensive theoretical and experimental analysis. Experiments were conducted on five datasets with 32 descriptor variations, eight drift detectors, and a non-detection context, resulting in 1440 combinations. The findings reveal three key conclusions: (i) reaction strategies must be tailored to different types of drift; (ii) severity, recurrence, and frequency descriptors have the highest impact, whereas speed has minimal influence; and (iii) there is a need to incorporate mechanisms for describing concept drift into the strategies designed to address it.