Navegando por Data de Publicação, começando com "2024-09-05"

Filtrar resultados por ano ou mês

Agora exibindo 1 - 9 de 9

Comparação de modelos de redes neurais na segmentação de vasos sanguíneos em imagens médicas
(Universidade Federal de São Carlos, 2024-09-05) Calvo, Ivan Duarte; Comin, Cesar Henrique; https://lattes.cnpq.br/9563440403120931; https://orcid.org/0000-0003-1207-4982; https://lattes.cnpq.br/5545908640929244; https://orcid.org/0009-0002-1457-2475
The analysis of blood vessels is responsible for extracting various important pieces of information in the healthcare field, and the impact of more precise analyses on disease studies and diagnoses holds positive potential. However, performing these analyses manually involves significant time and resource consumption. The segmentation of blood vessels in images represents a large part of this difficulty and cost. Technological advancements have enabled the implementation of machine learning techniques to carry out this crucial task, particularly through the use of neural networks, which has marked a significant evolution in the field. This work analyzes and compares various neural network models that have recently garnered significant attention, such as ResNet and EfficientNet, for example. The goal of this analysis and comparison is to obtain important insights into the advantages and disadvantages of applying each investigated model, such as situations where the models have lower accuracy and the influence of preprocessing and training stages on the models. The impact of varying the size and number of parameters in the models is also investigated, as smaller models may deliver very satisfactory results while consuming fewer computational resources and much less time. Finally, the tests indicated a good result with the models implemented with U-Net, especially RegNet and DenseNet. Also noteworthy were the results obtained by smaller models, like RegNetY_002, for instance, which in several cases achieved similar results in IoU and clDice as models with approximately five times more parameters.
Óleo essencial de citros no controle de Diaphorina citri Kuwayama
(Universidade Federal de São Carlos, 2024-09-05) Campos, Sabrina dos Santos; Devite, Fernando Trevizan; https://lattes.cnpq.br/4867480557280424; Conceição, Patrícia Marluci da; https://lattes.cnpq.br/6623193710607120; https://lattes.cnpq.br/2313134529464083
Previous studies have demonstrated the effectiveness of citrus essential oils (EOs) in pest control due to the effects of their components. Thus, the present study aimed to evaluate the efficacy of essential oils extracted from different citrus species, such as lemon (Citrus limon), orange (Citrus sinensis), and Tahiti lime (Citrus latifolia), in controlling the citrus psyllid (Diaphorina citri), the main vector of the disease Huanglongbing (HLB). The focus of this study was to investigate the action of these EOs at different concentrations (0.0%, 0.5%, 1.5%, 3%, and 6%) and modes of application (preventive and curative) against the psyllid. The experiment consisted of two trials, both conducted in a greenhouse. The first trial used EOs extracted from the Eureka and Amber lemon genotypes, while the second evaluated the EOs from orange and Tahiti lime, provided by Cisol do Brasil®. The trials were carried out in a completely randomized design with three replications, in a factorial scheme of 2x5x2 (two genotyp
Uma abordagem experimental do efeito estufa no ensino médio utilizando internet das coisas
(Universidade Federal de São Carlos, 2024-09-05) Raffa, Rodrigo Felipe; Soares, Antônio Augusto; https://lattes.cnpq.br/8093480629202515; Cantão, Renato Fernandes; https://lattes.cnpq.br/3016268479248046; https://lattes.cnpq.br/0679169959816286; https://orcid.org/0000-0001-5055-486X
This work aims to propose a low-cost experimental activity using Internet of Things (IoT) technology to address the greenhouse effect in Physics education. The study seeks to develop scientific and technological skills and competencies in high school students, in accordance with the National Common Curricular Base (BNCC). The proposed methodology uses the ESP32 microcontroller as a tool for data collection on the greenhouse effect, aiming to foster understanding and reflection on the causes and consequences of this phenomenon, as well as to stimulate the formation of a critical and responsible environmental awareness. The proposed experimental activity integrates theory and practice, allowing students to have an interactive and applicable learning experience in their daily lives. The use of the ESP32 enables the creation of a real-time monitoring system of the variables involved in the greenhouse effect, focusing on temperature. With this, students can visualize and analyze the collected data, understanding more concretely the impacts of the greenhouse effect. The development of the experimental activity was guided by the need to meet the BNCC guidelines, which emphasize the importance of scientific and technological education in high school. The proposed approach seeks to develop practical skills and stimulate critical thinking in students. The integration of IoT in Physics education allows the creation of more dynamic learning environments, where students can explore scientific concepts in a practical and contextualized manner. The use of emerging technologies, such as the ESP32, contributes to the formation of an education more connected with contemporary demands and the needs of an increasingly technological society. The expected results include improving students' understanding of scientific concepts related to the greenhouse effect and increasing engagement in sustainability issues. The proposed experimental activity aims to demonstrate that it is possible to teach Science in an innovative and effective way, using accessible and low-cost technological resources. The dissertation concludes that the implementation of experimental activities based on IoT can significantly contribute to scientific education in high school, preparing students to face environmental and technological challenges of the future.
Estudo comparativo entre replicação e erasure coding no armazenamento de dados em nuvem
(Universidade Federal de São Carlos, 2024-09-05) Silva, Sara Ferreira Bento da; Guardia, Hélio Crestana; https://lattes.cnpq.br/1780902767520967; https://orcid.org/0000-0001-5010-747X; https://lattes.cnpq.br/7936781650637000
The amount and diversity of digitally generated data has increased exponentially, leading to the increasing use of cloud storage services due to their scalability, high performance, and convenience. However, storing large volumes of data poses significant challenges to the infrastructure, especially in terms of financial costs and performance. This study compares two data protection methods used in cloud storage systems: replication and erasure coding. Replication creates copies of the data in different hardware locations, aiming to increase availability and reduce bandwidth during transmission, but results in high storage costs. In contrast, erasure codes offer lower storage costs by fragmenting the data, and are widely adopted in large-scale systems such as Microsoft Azure Storage and Google Cloud Storage. However, they present a high computational cost due to encoding and decoding operations. This study investigates the performance of OpenStack Swift with different erasure codes, analyzing how replication and erasure coding policies affect read and write operations and space usage. The results indicate that 2-copy replication is more efficient for small files, while 3-copy replication excels for larger files. On the other hand, erasure codes with (4,2) configuration show better efficiency in terms of storage space, with performance comparable to or superior to replication under certain conditions. The detailed analysis of the experimental results produced provides insights for choosing the most appropriate storage policy, considering different contexts and needs.
Automação de implantação de ambiente de computação em nuvem
(Universidade Federal de São Carlos, 2024-09-05) Dantas, Gabriel de Jesus; Guardia, Hélio Crestana; https://lattes.cnpq.br/1780902767520967; https://orcid.org/0000-0001-5010-747X
Cloud computing is a frequently used technology today, but one which is very complex in the process of implementing, stabilizing and maintaining environments (clusters) available to users. For OpenStack-based environments, this process results from a constant flow of human interaction, ranging on average from 2 to 4 hours of processing in average cases, without interference and debugging, and can take days in the worst cases, taking into account the processing done by the machine in this environment. In order to facilitate the implementation flow of a cloud infrastructure, the automation of this process is considered, aiming to reduce human interaction as much as possible, sending the installation, configuration and communication of software used in the cloud computing environment as an automatic task to the machine. Using Application Programming Interfaces (APIs) and software libraries for managing bare metal servers and Virtual Machines, and an orchestration engine, the goal is to transform the process of connecting and completely configuring the cloud environment into an automated script. As a result, this work generated strategies for building an API for deploying a cloud computing environment, strategies for adding new computers to a stable, functioning cloud environment, and a validation script that, when executed, configures a complete OpenStack infrastructure on a controller machine that initially has only an operating system and some type of Internet access.
Fundos offshore no Brasil: Uma análise entre 2006 e 2023
(Universidade Federal de São Carlos, 2024-09-05) Souza, Giulia Mateus de; Carrara, Aniela Fagundes; https://lattes.cnpq.br/2348376949393246
Considering the significance of personal investments in Brazil, this study aims to examine the landscape of offshore funds in Brazil between 2006 and 2023, with the objective of gathering economic data on this type of investment and understanding the amount of net assets they manage, as well as the possible relationship with the country's macroeconomic environment. To achieve this, a comparison with fixed-income funds, which are widely traditional in Brazil, was employed. The methodology used was descriptive data analysis, incorporating statistical measures and stationarity tests to understand the behavior of the analyzed net asset time series. The results suggest that offshore funds exhibit distinct dynamics compared to fixed-income funds, as offshore funds appear to carry a higher degree of randomness in the movement of their net assets.
Análise das métricas SMART em HDDs para modelos de predição de falha nos discos usados em data centers para a nuvem
(Universidade Federal de São Carlos, 2024-09-05) Baleeiro, Pedro Freire; Guardia, Hélio Crestana; https://lattes.cnpq.br/1780902767520967; https://orcid.org/0000-0001-5010-747X; https://lattes.cnpq.br/5189907384996909
Cloud data storage is one of the main activities of large cloud infrastructure providers. Objects (unstructured data) make up a large part of what is stored, and Hard Drives (HDs) are still the most widely used media to achieve significant amounts of storage at the best cost-benefit. The entire cloud service infrastructure must be redundant so that the data and the mechanism as a whole are reliable, durable, and available to the maximum. Thus, a significant aspect that has not yet been sufficiently explored and which is the subject of this work is the predictive maintenance of hard drives. In this work, the mechanisms behind this subject are explained by comparing redundancy strategies (Replication and Erasure Coding), monitoring SMART metrics and system reliability, calculating the Lifetime Value (RUL) of devices, and developing Machine Learning models to predict imminent failures in disks of a model. Furthermore, an experiment described in the article “Interpretable predictive maintenance for hard drives” [1] is replicated for a different disk model and range. Finally, considerations are gathered about the effectiveness of this model to treat the data in question and provide greater durability and reliability for a storage system based on hard drives. As a result, we seek to provide a broad knowledge of the subjects involved in this work.
Métodos de estimação de modelos de mistura para dados com Distribuição Poisson
(Universidade Federal de São Carlos, 2024-09-05) Almeida, Claudio Henrique Leão de; Milan, Luis Aparecido; https://lattes.cnpq.br/7435391829973844; https://orcid.org/0000-0001-6466-7865; https://lattes.cnpq.br/4118501720281480
Mixture models are only used when population data can be partitioned into subpopulations. This methodology allows the use of multiple probability distributions, so that each one determines the behavior of each subpopulation. In this work we study mixture model estimation methods for contagion data, focusing on the Bayesian approach. Two methods are presented here: EM (expectation-maximization algorithm), MH (Metropolis-Hasting). The first mentioned is based on maximum likelihood, there is no Bayesian inference. Applications were made using the EM and MH methods, in simulated databases with even variables. The methodologies are also applied to a real database. From two results, there are possible indications that the methods will perform well when the parameters are close. These estimates are even better for distant parameters. I also verified that as the sample size increases, these estimates are smaller, or what was expected.
Compreensão das práticas de ensino de ciências experimentais por professores dos anos iniciais do ensino fundamental de uma rede municipal: uma análise do ponto de vista dos saberes docentes
(Universidade Federal de São Carlos, 2024-09-05) Esteter, Querina Costa; Silva, João Ricardo Neves da; https://lattes.cnpq.br/5821775579111920; https://orcid.org/0000-0003-1311-0199; https://lattes.cnpq.br/6529531208830455; https://orcid.org/0009-0003-6060-1959
This study aims to understand how Science teaching practices occur in the initial years of Elementary School and what difficulties are encountered by the different elements that make up the educational scenario. To this end, the speeches of teachers and managers in different functions in a municipal network in a city in the interior of São Paulo were analyzed, in order to verify which knowledge is mobilized by teachers in the development of content in this area, seeking specific objectives to analyze the elements of teaching knowledge present about the need and practices of teaching natural sciences, characterize the elements of teaching knowledge present in the reports on the difficulties and possibilities of practices and characterize the relationship between the categories of knowledge expressed and the teaching system, including the municipality's official curriculum and the continuing teacher training process. To meet the proposed objectives, the study is characterized as descriptive and exploratory, using as a research strategy the analysis of semi-structured interviews, which were carried out with 12 professionals, teachers and professionals from the school management team and the data was assessed through content analysis, as presented by Bardin (1977). From this analysis, the elements were coded and organized into four categories, namely: inside Science teaching, teacher training in Science teaching, planning in Science teaching and the involvement of emotions. Furthermore, the data revealed the importance of interactions between the elements participating in the educational process and the curriculum, as a way of guiding the development of learning.