Programa de Pós-Graduação em Engenharia Elétrica - PPGEE/ITEC

URI Permanente desta comunidadehttps://repositorio.ufpa.br/handle/2011/2314

O Programa de Pós-Graduação em Engenharia Elétrica (PPGEE) do Instituto de Tecnologia (ITEC) da Universidade Federal do Pará (UFPA) foi o primeiro e é considerado o melhor programa de pós-graduação em Engenharia Elétrica da Região Amazônica. As atividades acadêmicas regulares dos cursos de mestrado e doutorado são desenvolvidas principalmente nas Faculdades de Engenharia Elétrica e Engenharia de Computação, supervisionadas pela Coordenação do Programa de Pós-Graduação em Engenharia Elétrica (CPPGEE).

Navegar

Agora exibindo 1 - 20 de 79

Acesso aberto (Open Access)
5G MIMO and LIDAR data for machine learning: mmWave beam-selection using deep learning
(Universidade Federal do Pará, 2019-08-29) DIAS, Marcus Vinicius de Oliveira; KLAUTAU JÚNIOR, Aldebaro Barreto da Rocha; http://lattes.cnpq.br/1596629769697284
Modern communication systems can exploit the increasing number of sensor data currently used in advanced equipment and reduce the overhead associated with link configuration. Also, the increasing complexity of networks suggests that machine learning (ML), such as deep neural networks, can effectively improve 5G technologies. The lack of large datasets make harder to investigate the application of deep learning in wireless communication. This work presents a simulation methodology (RayMobTime) that combines a vehicle traffic simulation (SUMO) with a ray-tracing simulator (Remcom’s Wireless InSite), to generate channels that represents realistic 5G scenarios, as well as the creation of LIDAR sensor data (via Blensor). The created dataset is utilized to investigate beam-selection techniques on vehicle-to-infrastructure using millimeter waves on different architectures, such as distributed architecture (usage of the information of only a selected vehicle, and processing of data on the vehicle) and centralized architectures (usage of all present information provided by the sensors in a given moment, processing at the base station). The results indicate that deep convolutional neural networks can be utilized to select beams under a top-M classification framework. It also shows that a distributed LIDAR-based architecture provides robust performance irrespective of car penetration rate, outperforming other architectures, as well as can be used to detect line-of-sight (LOS) with reasonable accuracy.
Acesso aberto (Open Access)
Abordagem Inteligente com Combinação de Características Estruturais para Detecção de Novas Famílias de Ransomware
(Universidade Federal do Pará, 2024-03-22) MOREIRA, Caio Carvalho; SALES JUNIOR, Claudomiro de Souza de; País de Nacionalidade Brasi
Ransomware is a malicious software that aims to encrypt user files and demand a ransom to unlock them. It is a cyber threat that can cause significant financial damage, as well as compromise privacy and data integrity. Although signature-based detection scanners commonly combat this threat, they fail to identify unknown ransomware families (variants). One method to detect new threats without the need to execute them is static analysis, which inspects the code and structure of the software, along with classification through intelligent approaches. The Detection of New Ransomware Families (DNFR) can be evaluated in a realistic and challenging scenario by categorizing and isolating families for training and testing. Hence, this thesis aims to develop an effective static analysis model for DNFR, which can be applied in Windows systems as an additional security layer to check executable files upon receipt or before execution. Early ransomware detection is essential to reduce the likelihood of a successful attack. The proposed approach comprehensively analyzes executable binaries, extracting and combining various structural features, and distinguishes them between ransomware or benign software employing a soft voting model comprising three machine learning techniques: Logistic Regression (LR), Random Forest (RF), and eXtreme Gradient Boosting (XGB). Results for DNFR demonstrated an average accuracy of 97.53%, precision of 96.36%, recall of 97.52%, and F-measure of 96.41%. Additionally, scanning and predicting individual samples took an average of 0.37 seconds. This performance indicates success in quickly identifying unknown ransomware variants and adapting the model to the constantly evolving landscape, suggesting its applicability in antivirus protection systems, even on resource-limited devices. Therefore, the method offers significant advantages and can assist developers of ransomware detection systems in creating more resilient, reliable, and fast-response solutions.
Acesso aberto (Open Access)
Agrupamento de fornos de redução de alumínio utilizando os algoritmos Affinity Propagation, Mapa auto–organizável de Kohonen (som), Fuzzy C–Means e K–Means
(Universidade Federal do Pará, 2017-10-11) LIMA, Flávia Ayana Nascimento de; CARDOSO, Diego Lisboa; http://lattes.cnpq.br/0507944343674734; OLIVEIRA, Roberto Célio Limão de; http://lattes.cnpq.br/4497607460894318
The continuous development of technology accounts for measures that provide industries benefits to grant them profitability and competitive advantage. In the mineralogy field, aluminum smelting usually requires substantial number of cells, also known as reduction pots, to produce aluminum in a continuous and complex process. Analytical monitoring is essential for those industries’ competitive advantage, given that during operation some cells show behavior similar to others, thereby forming clusters of cells. These clusters depend on data patterns usually implicit or invisible for the operation, but can be found by data analysis techniques. In this work four clustering techniques are presented to that end: the Affinity Propagation; the Kohonen Self Organizing Map; the Fuzzy C–Means; and the K–Means Algorithm. These techniques are used to find and group cells that share similar behavior, by analysing seven variables which are closely related to the aluminum reduction process. This work aims at addressing the benefits of clustering, especially by simplifying the aluminum potline analysis, once a large group of cells might be summarized in one sole group, what can provide more compact yet rich information for data driven modeling and control. Moreover, the identification of similar data patterns in clusters makes the task of those who is going to be in charge of analyzing these dats. This work also identifies the ideal cluster size for each technique applied.
Acesso aberto (Open Access)
Algoritmo memético cultural para otimização de problemas de variáveis reais
(Universidade Federal do Pará, 2019-03-29) FREITAS, Carlos Alberto Oliveira de; SILVA, Deam James Azevedo da; http://lattes.cnpq.br/8540875293894747; OLIVEIRA, Roberto Célio Limão de; http://lattes.cnpq.br/4497607460894318
Technology has made great strides in recent years, but computing resources for certain applications need optimization so that the costs involved in solving some problems are not high. There is a very broad area of research for the development of efficient algorithms for multimodal optimization problems. In the last two decades the use of evolutionary algorithms in multimodal optimization has been shown to be a success. Among these evolutionary algorithms, which are global search algorithms, one can cite the use of Cultural Algorithms. A natural enhancement of the Cultural Algorithm is its hybridization with some other local search algorithm, so as to have the advantages of global search combined with local search. However, the local search Cultural Algorithms used for multimodal optimization are not always evaluated by efficient statistical tests. The objective of this work is to analyze the behavior of the Cultural Algorithm, with populations evolved by the Genetic Algorithm, when the local search heuristics are used: Tabu Search, Beam Search, Climbing and Simulated Annealing. One of the contributions of this work was the updating of the topographic knowledge of the cultural algorithm by the use of the triangular area defined by the best results found in the local search. To perform the analysis, a memetic algorithm was developed by hybridizing the cultural algorithm with the local search heuristics mentioned, being selected one at a time. Real world problems usually have multimodal characteristics, so the evaluations were performed using multimodal benchmark functions, which had their results evaluated by non-parametric tests. In addition, the memetic algorithm was tested on real optimization problems with constraints in the engineering areas. In the evaluations carried out, the developed Cultural Algorithm presented better results when compared to the available results of the researched scientific literature.
Acesso aberto (Open Access)
Alocação de dois níveis para uma arquitetura h-cran baseada em offloading
(Universidade Federal do Pará, 2019-01-24) GONÇALVES, Mariane de Paula da Silva; BARROS, Fabrício José Brito; http://lattes.cnpq.br/9758585938727609; CARDOSO, Diego Lisboa; http://lattes.cnpq.br/0507944343674734
The accelerated data and apps growth represents significant challenges to the next generation of mobile networks. Amongst them, it is highlighted the necessity for a co-existence of new and old patterns during the transition of architectures. Thus, this paper has investigated solutions for offloading into a hybrid architecture, also known as H-CRAN (Heterogeneous Cloud Radio Access Network Architecture), that centralizes processing and searches a better use of the network resources. The strategy of optimization was analyzed through the evolutive algorithm PSO (Particle Swarm Optimization), in order to find a suboptimal solution to the allocation of two levels (TLA) in the H-CRAN architecture and another one based on FIFO (First In, First Out), for benchmarking purposes. SNR (Noise Interference Signal) average, Maximum Bit Rate, the number of users with or without connections and number of connections in RRHs and macro were used as performance measurements. Through the results, it was noticed an improvement of approximately 60% in the Maximum Bit Rate when compared to the traditional approach, enabling a better service to the users.
Acesso aberto (Open Access)
Alocação ótima de geração distribuída em redes de distribuição utilizando algoritmo híbrido baseado em cuckoo search e algoritmo genético
(Universidade Federal do Pará, 2018-09-02) OLIVEIRA, Victoria Yukie Matsunaga de; AFFONSO, Carolina de Mattos; http://lattes.cnpq.br/2228901515752720
This thesis presents a novel Cuckoo Search (CS) algorithm called Cuckoo-GRN (Cuckoo Search with Genetically Replaced Nests), which incorporates the benefits of genetic algorithm (GA) into the CS algorithm. The proposed method handles the abandoned nests from CS more efficiently by genetically replacing them, significantly improving the performance of the algorithm by establishing optimal balance between diversification and intensification. The algorithm is used for the optimal location and size of distributed generation units in a distribution system, in order to minimise active power losses while improving system voltage stability and voltage profile. The allocation of single and multiple distribution generation units is considered. The proposed algorithm is extensively tested in mathematical benchmark functions as well as in the 33-bus and 119-bus distribution systems. Simulation results show that Cuckoo-GRN can lead to a substantial performance improvement over the original CS algorithm and others techniques currently known in literature, regarding not only the convergence but also the solution accuracy.
Acesso aberto (Open Access)
Análise de desempenho de algoritmos para classificação de sequências representando faltas do tipo curto-circuito em linhas de transmissão de energia elétrica
(Universidade Federal do Pará, 2019-12-05) FREIRE, Jean Carlos Arouche; MORAIS, Jefferson Magalhães de; http://lattes.cnpq.br/5219735119295290; CASTRO, Adriana Rosa Garcez; http://lattes.cnpq.br/5273686389382860
Maintaining power quality in electrical power systems depends on addressing the major disturbances that may arise in their generation, transmission and distribution. Within this context, many studies have been developed aiming to detect and classify short circuit faults in electrical systems through the analysis of the electrical signal behavior. Transmission line fault classification systems can be divided into two types: online and post fault classification systems. In the post-missing scenario the signal sequences to be evaluated for classification have variable length (duration). In sequence classification it is possible to use conventional classifiers such as Artificial Neural Networks, Support Vector Machine, K-nearest neighboors and Random forest. In these cases, the classification process usually requires a sequence preprocessing or a front end stage that converts the raw data into sensitive parameters to feed the classifier, which may increase the computational cost of the classification system. An alternative to this problem is the FBSC-FrameBased-Sequence Classification (FBSC) architecture. The problem with FBSC architecture is that it has many degrees of freedom in designing the model (front end plus classifier) and it should be evaluated using a complete dataset and rigorous methodology to avoid biased conclusions. Considering the importance of using efficient short-circuit fault classification methodologies and mainly with low computational cost, this paper presents the results of the KNN-DTW (K-Nearest Neighbor) algorithm analysis study associated with Dynamic similarity measurement. Time Warping (DTW) and HMM (Hidden Markov Model) algorithm for fault classification task. These two techniques allow the direct use of data without the need for front ends for signal pre-processing, as well as being able to handle multivariate and variable time series, such as signal sequences for the post-miss case. To develop the two proposed systems for classification, simulated data of short-circuit faults from the UFPAFaults public database were used. To compare results with methodologies already presented in the literature for the problem, the FBSC architecture was also evaluated for the same database. In the case of FBSC architecture, different front ends and classifiers were used. The comparative assessment was performed from the measurement of error rate, computational cost and statistical tests. The results showed that the HMM-based classifier was more suitable for the problem of classification of short circuits on transmission lines.
Acesso aberto (Open Access)
Análise do QoE para streaming de vídeo em redes aéreas 5G mmWave de conectividade dupla
(Universidade Federal do Pará, 2022-02-18) PEREIRA, Ronilson Williame da Silva; SILVA, Marcelino Silva da; http://lattes.cnpq.br/7080513172499497
The evolution in mobile telecommunications has allowed the emergence of new network formats aimed at meeting the growing demand for wireless data, mainly fueled by more viewing of video content. The scarce spectrum available in current cellular networks does not seem to be able to meet this explosion of wireless data, motivating a shift to explore new frequency bands. Currently, one of the major problems of current mobile networks is congestion in denser areas such as concerts, sporting events, and festivals (temporary events), as we have a greater concentration of users, consequently increasing the volume of data traffic. Millimeter-wave (mmWave) communication, frequency bands between 30 and 300 GHz emerges as an essential part for the next generation of fifth-generation cellular networks (5G), as they adopt higher carrier frequencies offering high bandwidth and lower latency. However, system performance degrades due to high propagation loss and the sensitivity of the links to obstacles. To mitigate such factors, this dissertation proposes a dual connectivity 5G mmWave architecture using LTE (Long Term Evolution) and the use of UAVs (Unmanned Aerial Vehicles) as airbase stations to provide connectivity guarantees and quality of experience. in video streaming, one scenario where the proposal might be a suitable solution is to provide coverage for temporary events. Through simulations, an analysis of the Quality of Experience (QoE) was presented about the transmission of 4K videos in order to evaluate the transmission and quality of videos perceived by users. The simulated results show the efficiency of the system for multimedia applications using videos, improving the QoE of wireless users by 43%.
Acesso aberto (Open Access)
Análise dos fatores relacionados ao desempenho das escolas no IDEB: estudo de caso no Estado do Pará
(Universidade Federal do Pará, 2022-02-11) GOMES, Vitor Hugo Macedo; SILVA, Marcelino Silva da
The complexity of identifying all the factors that are related to the performance of schools on the Basic Education Development Index (IDEB) is enormous. In this study, three databases were analyzed with the objective of identifying several factors that correlate with low performance in state schools in the state of Pará. Initially, it was observed through the analysis that 142 municipalities in the state were at risk of not meeting the goal regarding the reduction of school dropouts and, consequently, affecting the performance of schools. This study used educational data mining techniques to, first, select variables with structural characteristics in the teaching environment, comparing the schools with higher and lower performance in IDEB, identifying possible relationships with school dropouts. Then, the Randon Florest (RF) algorithm was used to select the most important variables that directly or indirectly impact the IDEB index. After the selection phase, the variables were submitted to the Linear Regression (LR) algorithm. The results reveal that in the group of schools below average in IDEB, 60.6% reside in families with incomes up to one minimum wage, while 37.5% have incomes above one minimum wage. In the group of schools above average in IDEB, 42.4% live in families with incomes up to one minimum wage, while 51.6% live in families with incomes above one minimum wage. Evidencing that family income is related to better IDEB scores and, consequently, better infrastructure conditions. The results also indicate that the income of students’ families is related to the average family income in the analyzed municipalities. Next, variables related to parents’ income were used to identify a possible relationship between parents’ schooling and students’ performance. Finally, the analysis ends with the analysis of the impact of the Municipal Human Development Index (HDI) on the variables related to the students’ grades, the teachers’ qualifications, and the teachers’ experience in the school environment. The results reveal that there is a correlation between the index and student learning in the classroom. On the other hand, better IDEB scores are directly related to the adequacy of the curriculum to the subject taught, in addition to good working conditions for teachers.
Acesso aberto (Open Access)
Análise e classificação de severidade de COVID-19 usando aprendizado de máquina
(Universidade Federal do Pará, 2022-08-16) LIMA, Marco Antonio Loureiro; CARDOSO, Diego Lisboa; http://lattes.cnpq.br/0507944343674734
In the last years, with the alarming growth of COVID-19 cases, a highly contagious viral disease, new forms of diagnosis and control for this sickness have become necessary to the spread decreases until the population is effectively vaccinated. In this context, Artificial Intelligence (AI) and its subfields appear as possible alternatives to help and provides a response to combat the virus. Some Machine Learning (ML) methods are shown as an answer to control this disease, these methods can perform an analysis based on a set of symptoms presented by the patient and consequently indicating the diagnosis, as well as streamline the treatment process. To achieve this goal in this paper, three models that uses ML methods to predict COVID-19 severity on different degrees are proposed, unlike other works whose purpose was to diagnose only the presence or absence of COVID-19, this paper aims to improve the classification of the patient’s disease state. The results in each of these models are evaluated through the metrics established in this work. Furthermore, there are distinct suggestions to improve the analysis and make predictions with greater accuracy..
Acesso aberto (Open Access)
Uma análise sócio-demográfica da incidência de hanseníase na Amazônia legal brasileira: abordagem baseada em redes bayesianas
(Universidade Federal do Pará, 2019-02-08) GOMES, José Maria da Silveira; FRANCÊS, Carlos Renato Lisboa; http://lattes.cnpq.br/7458287841862567
Leprosy is a millenarian contagious disease, with chronic and stigmatizing characteristics, from the remotest times of humanity until today. It is characterized as a disease of the poor and Brazil is the second country in the world with the highest incidence. The lack of public policies aimed at reducing poverty through the improvement of socio-economic factors in the country is directly related to the incidence of the disease in Brazil. Strategies for control and monitoring should follow intelligent actions. One of the solutions for monitoring the disease is the use of Bayesian networks as a probabilistic method for taking decisions on both the control and the procedures to adopt in order to reduce the incidence of the disease. The objective of the present study is to analyse the association of leprosy incidence in relation to indicators of human development, habitation and income level, considering the Brazilian Amazon region in relation to the entire country. An ecological study, based on data obtained on cases of leprosy in Brazil for the year 2010, obtained from the Information System of Hardship Notifications (SINAN) through the Informatics Department of the National Health Service (DATASUS) and the socio-economic indicators found in the Demographic Census Research database of the Brazilian Institute for Geographical and Statistical Survey – IBGE, as well as information from the Municipal Human Development Index, regarding education and income, obtained from the website of the Human Development Atlas of Brazil, also for the year 2010. The methodology combined data mining with the analysis of spatial distribution. The Bayesian network technique was used aimed at measuring the association between variables of the domain of the problem as well as to establish the analogy of the data between the municipalities under study with data for all other Brazilian municipalities. Applying the algorithm K2 relevant associations were found for the following indicators applied in the investigation: Brazilian Legal Amazon, Municipal Human Development Index of Income and Education and Household Housing Condition. Using the Bayesian network model adopted, there is a significant association between the percentage of homes with more than 2 inhabitants and the rate of incidence of leprosy. Although the relationship between the rate of incidence, socio-economic factors (no water supply, no toilet, poverty and overcrowding of the home), low educational indices and income has already been reported in several studies, the insertion of the indicators that considers population density of the home was a novel proposition of the present study and the indicators of greatest most significance of this investigation. The analysis of leprosy incidence with respect to spatial distribution, comparing the Amazon region with the entire country, revealed that public policies for habitation in the studied region were almost non-existent, since the population density of homes is very high, facilitating the appearance of contagious diseases such as leprosy.
Acesso aberto (Open Access)
Aplicação de fanets e ca-markov para captura de imagens para o estudo de uso e cobertura da terra em projetos de assentamentos na amazônia
(Universidade Federal do Pará, 2019-12-06) SOUZA, Jorge Antonio Moraes de; FRANCÊS, Carlos Renato Lisboa; http://lattes.cnpq.br/7458287841862567
Projetos de assentamentos de reforma agrária são uma das medidas adotadas pelo governo na tentativa de criar um relacionamento sustentável com a natureza. Como a área de assentamentos cobre mais de 77.483.317,86 hectares da Amazônia Legal, é essencial compreender as causas da degradação ambiental desses espaços. Isto posto, foram utilizados, de forma combinada, cadeias de Markov e autômatos celulares (CA-Markov) para, a partir de duas imagens classificadas, prever cenários de mudanças no uso e cobertura da terra (LULC). Esta tese apresenta uma metodologia inovadora que difere daquelas usualmente utilizadas em CA-Markov, pois os aspectos de tempo e espaço são observados pela cadeia de Markov e servem como base para a função de transição do autômato celular (CA). A metodologia também contempla a aquisição de imagens, nesse sentido, como a região de interesse permanece, em boa parte do ano, com uma cobertura de nuvens significativa, a obtenção de imagens por sensores ópticos, fica prejudicada, por conta disso, foi imperativa a busca por uma alternativa. As Flying Ad-hoc Networks (FANETs) podem ser utilizadas para complementar informações da região de estudo, capturando imagens de alta qualidade, sem o inconveniente das nuvens. Por outro lado, os nós da rede precisam manter, pelo maior tempo possível, a conexão entre eles, o que é dificultado pela mobilidade e autonomia de voo dos drones. Por esse motivo, é imprescindível a utilização de um protocolo de roteamento que seja capaz de adaptar-se à dinâmica da rede. Além disso, também foi desenvolvido um algoritmo de roteamento baseado em sistema Fuzzy. Testes e simulações foram realizadas com o intuito de validar tanto a metodologia geral MAPS, quanto o protocolo de roteamento.
Acesso aberto (Open Access)
Aplicação de redes neurais artificiais para predição de RSSI e SNR em ambiente de bosque amazônico
(Universidade Federal do Pará, 2024-06-11) BARBOSA, Brenda Silvana de Souza; ARAÚJO, Jasmine Priscyla Leite de; http://lattes.cnpq.br/4001747699670004; https://orcid.org/0000-0003-3514-0401; BARROS, Fabrício José Brito; http://lattes.cnpq.br/9758585938727609
The presence of green areas in urbanized cities is crucial to reduce the negative impacts of urbanization. However, these areas can influence the signal quality of IoT devices that use wireless communication, such as LoRa technology. Vegetation attenuates electromagnetic waves, interfering with data transmission between IoT devices, resulting in the need for signal propagation modeling that considers the effect of vegetation on its propagation. In this context, this research was conducted at the Federal University of Pará, using measurements in a wooded environment composed of the Pau-Mulato species, typical of the Amazon. Two propagation models based on machine learning, GRNN and MLPNN, were developed to consider the effect of Amazonian trees on propagation, analyzing different factors such as the height of the transmitter relative to the trunk, the beginning of the foliage, and the middle of the tree canopy, as well as the LoRa spreading factor (SF) 12 and the copolarization of the transmitter and receiver antennas. The best models were the machine learning ones, GRNN and MLPNN, which demonstrated greater accuracy, achieving root mean square error (RMSE) values of 3.86 dB and 3.8614 dB, and standard deviation (SD) of 3.8558 dB and 3.8564 dB, respectively. On the other hand, compared to classical models in the literature, the best-performing model was the Floating Intercept (FI) model, with RMSE and SD errors around 7.74 dB and 7.77 dB, respectively, while the FITU-R model had the highest RMSE and SD errors, around 26.40 dB and 9.65 dB, respectively, for all heights and polarizations. Furthermore, the importance of this study lies in its potential to boost wireless communications in wooded environments, as it was observed that even at short distances at heights of 12 m and 18 m, the SNR (Signal-to-Noise Ratio) had lower values due to the influence of the foliage, but it was still possible to send and receive data. Finally, it was shown that vertical polarization achieved the best results for the Amazon forest environment.
Acesso aberto (Open Access)
Arquitetura de modelos híbridos, machine learning e otimizadores para análise de consumo de energia elétrica e produtividade em pintura automotiva
(Universidade Federal do Pará, 2024-03-27) OLIVEIRA, Rafael Barbosa de; OLIVEIRA, Roberto Célio Limão de; http://lattes.cnpq.br/4497607460894318
Strategies for optimizing energy consumption in the painting stages are emerging as key factors in promoting more sustainable and competitive production in the automotive sector. This dissertation seeks to predict energy consumption and maximize productivity in automotive painting, using an approach that combines variable selection, hybrid models, hyperparameters of these models and meta-heuristic optimization in a 3-stage architecture. Automotive painting processes have variables in the form of time series that describe the history of energy consumption. In stage 1, the best machine learning model is chosen (Random Forest, Long-Short Term Memory, XGBoost and GRU-LSTM) to predict energy consumption time series at t+1. In step 2, the RF, XGBoost and Dense Artificial Neural Network (ANN) models are evaluated to select the best predictor of the number of vehicles produced (cycles). In step 3, the best metaheuristic between Genetic Algorithm (GA), Differential Evolution (DE) and Particle Swarm Optimization (PSO) is selected to optimize the energy consumption predicted by the best model from step 1, using the best model from step 2 as a fitness measure. The final architecture reduced the energy consumed by up to 16% and increased the cycle by 127%, using the GRU-LSTM models in step 1, Dense ANN in step 2 and DE in step 3. The results highlight the opportunity to use the proposed approach to optimize energy consumption and productivity in automotive painting.
Acesso aberto (Open Access)
Uma arquitetura de pré-processamento para análise de sentimento em mídias sociais em português brasileiro
(Universidade Federal do Pará, 2018-08-23) CIRQUEIRA, Douglas da Rocha; SANTANA, Ádamo Lima de; http://lattes.cnpq.br/4073088744952858
The Web 2.0 and the evolution of Information Technologies have brought novel interaction and relationship channels. In this context, the Online Social Networks (OSN) are an example as platforms which allow interactions and sharing of information between people. In this scenario, it is possible to observe the adoption of OSN as a channel for posting opinions regarding products and experience. This scene presents an excellent opportunity for companies that aim to improve products, services and marketing strategies, given OSNs are powerful sources of massive unstructured data generated by consumers (UGC), with opinions and reviews concerning offers, in platforms such as Facebook, Twitter and Instagram. Brazil is a highlight in this scenario, where this phenomenon can be observed, as the Brazilian population is one of the most active in social media platforms in the world. This makes it a country full of opportunities to market exploitation. In this context, computational techniques of Opinion Mining and Sentiment Analysis (SA) are applied aiming to infer the polarity (positive, negative, neutral) regarding a sentiment associated to texts, and can also be applied in data from OSN to evaluate the feedback from a target audience. Although the existing diversity of SA strategies reported in the literature, there are still challenges faced in the application of SA in text data from OSN, given the characteristics of the language adopted in such platforms. The state of art is focused on SA towards the English language, and the existing proposals for Brazilian Portuguese do not have a standardized methodology for preprocessing steps. In this context, this research investigates an approach with no translation, and proposes a novel preprocessing architecture for SA towards Brazilian Portuguese, aiming to provide enriched features to SA algorithms. The proposal was compared with well-established baselines from the literature, and the obtained results indicate that this architecture can overcome the state of art recall in at least 3% , for 6 out of 7 datasets evaluated.
Acesso aberto (Open Access)
Ciência de Dados Aplicada em Dados Públicos: Estudos de Caso Acerca da Previdência Social Brasileira.
(Universidade Federal do Pará, 2020-04-17) FELIX JÚNIOR, Francisco Eguinaldo de Albuquerque; SILVA, Marcelino Silva da; http://lattes.cnpq.br/7080513172499497
Data Science is an interdisciplinary area related to data analysis, which aims to extract knowledge and possible decision-making about specific problems. In this context, open government data, which often need pre-treatments and computational methods to process their data sets, present themselves as potential sources of information to be explored taking the Data Science’s perspective, allowing the development of strategies each time more efficient and optimized in public management. Given this, and allied to the recent discussions related to the reform in the Brazilian social security, this dissertation presents two case studies referring to analyzes in the national social security system. The first study used the microdata referring to the demographic censuses of 2000 and 2010, made available by IBGE, proposing to evaluate the participation that retirements and pensions have in the income inequality of the population in the years evaluated about Brazilian states and municipalities. The results show that, although the analyzed benefits contribute to the Brazil income concentration, the portion corres=ponding to a minimum wage contributes to the deconcentration of income, and the portion above one salary contributes to the concentration, being a repetitive pattern throughout the country. On the other hand, the second study proposed an evaluation of the impacts caused by the pension reform, which is proposed in PEC 06/2019. It was observed that PEC 06/2019 would hinder access to benefits, in which approximately 83,28% of the pensions would not have been granted had it been in effect since 1995.
Acesso aberto (Open Access)
Classificação de arritmias cardíacas através de uma estrutura competitiva de redes neurais convolucionais autoassociativas
(Universidade Federal do Pará, 2023-05-11) CORRÊA FILHO, Sérgio Teixeira; CASTRO, Adriana Rosa Garcez; http://lattes.cnpq.br/5273686389382860
This work proposes a system for classifying cardiac arrhythmias based on a competitive structure of Autoassociative Convolutional Neural Networks. Three neural networks were trained to reconstruct Electrocardiogram (ECG) signals for cases of patients with supraventricular, ventricular and normal beats. After training, the networks were allocated in a competitive parallel structure for classification of arrhythmias. The MIT-BIH arrhythmia public database of ECG signals was used for training and testing the networks, and for each ECG signal, from each patient, the QRS complexes of the heartbeats were extracted, which were the characteristics used as input. for the system, and these signals, which were in the form of temporal signals (1D), were transformed into digital images (2D) in order to use the capacity of convolutional neural networks for pattern recognition and feature extraction in images. For the development and performance analysis of the proposed structure, two paradigms that have been used in works already presented in the literature were used: interpatient paradigm and intrapatient paradigm, and the system obtained an accuracy of 96.97%, sensitivity of 96.30% and precision of 93.59% for the intrapatient case and accuracy of 94.05%, sensitivity of 70.43% and precision of 65.74% for the interpatient case. A comparative analysis with results from arrhythmia classification systems already presented in the literature shows that the proposed system presented similar results or, in some cases, better results than those already obtained, thus showing the applicability of the proposed structure to the problem
Acesso aberto (Open Access)
Classificação de ransomware utilizando MLP, redução de dimensionalidade e balanceamento de classes
(Universidade Federal do Pará, 2023-07-03) PEREIRA, George Tassiano Melo; SALES JÚNIOR, Cloaudomiro de Souza de; http://lattes.cnpq.br/4742268936279649
Ransomware is a type of malware that prevents or limits user access to system and files until aransom is paid. Combating this threat is difficult due to its rapid spread and constant changes in the encryption techniques used. Machine learning algorithms such as Artificial Neural Networks have been touted as promising tools in classifying ransomware because they can learn to identify complex patterns and features in large amounts of data. This allows neural networks be trained on sample examples of malicious software, including ransomware, and then be able to classify new examples with high accuracy. Furthermore, neural networks are also capable of learning and adapting to changes in malware behavior, making them effective tools for detecting new types of ransomware. In this work, three types of ransomware classification by ANN are explored within a composite pipeline with dimensionality reduction by Kernel PCA and class balancing with the random oversampling approach. The MLP (Multi-layer Perceptron) reached an average of 98% accuracy in the binary classification and 85% accuracy in the goodware family classification, where such values surpass the previous results and thus demonstrate the effectiveness of the inclusion of the class balancing in improving the ransomware detection model.
Acesso aberto (Open Access)
Classificação de regiões de desmatamento via imagens do satélite landsat no nordeste do Pará
(Universidade Federal do Pará, 2023-12-18) CANAVIEIRA, Luena Ossana; COSTA, João Crisóstomo Weyl Albuquerque; http://lattes.cnpq.br/9622051867672434
Acesso aberto (Open Access)
Classificação de tumores cerebrais: um estudo comparativo entre rede neural convolucional e rede neural convolucional com mecanismo de atenção
(Universidade Federal do Pará, 2024-09-30) SILVA, Ulrich Kauê Mendes Alencar da; CASTRO, Adriana Rosa Garcez; http://lattes.cnpq.br/5273686389382860
Brain tumors are neurological diseases with a high potential impact on the lives of affected individuals, requiring a rapid and accurate diagnosis through complementary imaging tests, such as magnetic resonance imaging, which is considered the gold standard in this process. Considering the need for faster diagnosis, classification systems based on Machine Learning have been developed and within this context, this dissertation aims to present a comparative study between a Convolutional Neural Network (CNN) and a CNN with an attention mechanism, developed for the classification of brain tumors from magnetic resonance images. The comparative study aims to identify the impact of the attention mechanism on the performance of the CNN for tumor classification. For the development and evaluation of the proposed models, a public database was used, collected from the Kaggle website and made available by Masoud Nickparvar, which is composed of 7023 brain magnetic resonance images, segmented into four classes: glioma, meningioma, no tumor and pituitary. As a result, from the performance metrics obtained, considering the image base used for testing in both CNNs, an improvement in the CNN performance was observed after the introduction of the attention mechanism, where the network with this mechanism presented an increase of 1.98% in the accuracy metric, 2.07% in the precision metric, 2.18% in the sensitivity metric and 1.72% in the F1-score metric in relation to the CNN without the attention mechanism. It is also possible to highlight the results obtained in particular for the meningioma tumor class, since the CNN without the attention mechanism presented difficulties in classifying this class and, after the integration of the attention mechanism, the model obtained an accuracy increase of 6.54% for this class.