|
Thèses |
|
1
|
-
FERNANDA PEREIRA DOS SANTOS RODRIGUES
-
**
-
Leader : ALISSON MARQUES DA SILVA
-
MEMBRES DE LA BANQUE :
-
ALISSON MARQUES DA SILVA
-
ANDRE PAIM LEMOS
-
ELISANGELA MARTINS DE SA
-
FLAVIO VINICIUS CRUZEIRO MARTINS
-
Data: 15 févr. 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
This work introduces an evolving fuzzy approach constructed based on a non-supervised recursive clustering algorithm with participatory learning and multivariate Gaussian membership functions. From this approach, four models are introduced, called eFLS (evolving Fuzzy with Multivariable Gaussian Participatory Learning and Weighted Recursive Least Squares), eFMI (evolving Fuzzy with Multivariable Gaussian Participatory Learning and Multi-Innovations Recursive Weighted Least Squares), eFCE (evolving Fuzzy with Multivariable Gaussian Participatory Learning and Recursive Maximum Correntropy) and eFTLS (evolving Fuzzy with Multivariable Gaussian Participatory Learning and Recursive Weighted Total Least Squares). In the proposed approach, clusters are created using a compatibility measure and an alert mechanism. The compatibility measure is computed by Euclidian or Mahalanobis distance according to the number of samples in the cluster. Inactive clusters are excluded by an age and population based-method. Redundant clusters are merged whenever is a noticeable overlap between two clusters. The consequent parameters are updated by a recursive algorithm of Weighted Least Squares, Multi-Innovations Weighted Least Squares, Maximum Correntropy, or Weighted Total Least Squares. Finally, the performance of the models is evaluated and compared with alternative state-of-the-art models in time series forecasting and non-linear system identification problems. Computational experiments and comparisons suggest that the proposed models have superior or comparable performance to alternative models.
|
|
2
|
-
RAULIVAN RODRIGO DA SILVA
-
**
-
Leader : THIAGO MAGELA RODRIGUES DIAS
-
MEMBRES DE LA BANQUE :
-
THIAGO MAGELA RODRIGUES DIAS
-
ALISSON MARQUES DA SILVA
-
Jesús Pascual Mena Chalco
-
WASHINGTON LUIS RIBEIRO DE CARVALHO SEGUNDO
-
Data: 17 mars 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
The main objective of this work is to present an overview of Brazilian technical production based on the analysis of patents registered at the National Institute of Industrial Property (INPI), as well as international repositories such as Espacenet. Initially, using the curricular base of the Lattes Platform of the National Council for Scientific and Technological Development (CNPq), this study seeks to carry out an analysis of patents registered in the curricula of Brazilian researchers and their validation with the Espacenet repository, in order to verify the validity of the data, as well as obtaining a set of additional information about the recovered patents. For an analysis that extrapolates the data extracted from the résumés, complementary information will be extracted from the patents identified in the résumés through the collection of information in the INPI repository and also in the Espacenet repository, the latter being an international repository of patent registrations of recognized relevance . In order to identify innovation networks and the formation of invisible schools resulting from collaboration between researchers throughout the national territory, this analysis will be carried out using data extraction tools, text mining, graph construction, as well as algorithms of natural language processing. Subsequently, patterns will be explored and data quantification will be explored, as well as an evaluation of the formed collaboration networks, with an emphasis on individuals and institutions.
|
|
3
|
-
FELIPE DUARTE DOS REIS
-
**
-
Leader : HENRIQUE ELIAS BORGES
-
MEMBRES DE LA BANQUE :
-
BRUNO ANDRE SANTOS
-
ELIZABETH FIALHO WANNER
-
HENRIQUE ELIAS BORGES
-
ROGERIO MARTINS GOMES
-
Data: 25 mars 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
Solving complex optimization problems using meta-heuristics and their combinations are, in general, a hard approach, since a great knowledge of the problem often is necessary. An alternative to the development of new algorithms, or their manual hybridization, is to use the collaboration and communication mechanisms inherent in the modeling of multi-agent systems (MMAS). The D-Optimas architecture is an MMAS based on the actor model, where each agent encapsulates a different meta-heuristic, and with a learning mechanism they are allowed to collaborate in finding the best solution to an optimization problem. The agents interact in the search space that is divided into regions, which have an independent behavior, being able to receive new solutions, spliting into three new regions or merge with another one. However, the execution of the D-Optimas is limited to a small number of nodes in a cluster. Extending it, allowing its execution in a cluster with an unfixed number of nodes is essential for solving large-scale problems and extracting reliable data to analyze the architecture’s performance. Therefore, this work aims to consolidate the D-Optimas architecture from the point of view of a distributed system, fault tolerant, with load balancing and location transparency, making it resilient and horizontally scalable. The results obtained by this research so far shows that the architecture scales efficiently in a cluster with up to six nodes without losing performance. It is expected, with the continuation of this work, to study the dynamic hybridization of architecture with a wide variety of agents.
|
|
4
|
-
ALESSANDRA ANGELITA CARNEIRO ALVES
-
**
-
Leader : JOSE LUIZ ACEBAL FERNANDES
-
MEMBRES DE LA BANQUE :
-
JOSE LUIZ ACEBAL FERNANDES
-
ELIZABETH FIALHO WANNER
-
LUIS ALBERTO D AFONSECA
-
ANA PAULA PINTADO WYSE
-
Lillia dos Santos Barsante Silva
-
Data: 28 mars 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
Aedes aegypti and Aedes albopictus are the main vectors of arbovirus infections such as dengue (DENV), zika (ZIKV), chikungunya fever (CHIKV) and yellow fever (YFV) in the world, being responsible by the infection of millions of people annually. The range of the highest global incidence of these vectors is concentrated in tropical and subtropical regions due to the climate of these regions favoring the development of mosquitoes. Rainfall has a considerable influence on the development of Aedes, with increasing rainfall, the number of breeding sites available for egg laying increases, causing a positive correlation with the incidence of arboviruses. Mathematical models related to meteorological variables have been widely developed in order to assist in the control of the vector. The oviposition of Aedes spp. it can occur in wet and dry environments and this factor is able to modulate the population influencing its seasonal pattern by the female and the seasonal pattern of the vector population. In this work a theoretical-numerical study of the model will be carried out in order to reproduce the dependence of the entomological parameters of the mosquito’s life cycle with the accumulated weekly rainfall of certain Brazilian regions. An entomological compartmental mathematical model composed of three nonlinear differential equations representing the dynamics of egg populations in dry and wet environments, and the female population under the effect of accumulated weekly rainfall variables, was proposed and analyzed. The model was validated by comparison with experimental catch data obtained by entomological monitoring in the city of Caratinga, Lavras and Sete Lagoas (Minas Gerais, Brazil). It was found that the model describes the mosquito’s life cycle and is able to assist in the monitoring and control of Aedes.
|
|
5
|
-
TALITA GOMES SILVA
-
**
-
Leader : GIANCARLO QUEIROZ PELLEGRINO
-
MEMBRES DE LA BANQUE :
-
GIANCARLO QUEIROZ PELLEGRINO
-
JOSE GERALDO PEIXOTO DE FARIA
-
WESLLEY FLORENTINO DE OLIVEIRA
-
Data: 30 mars 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
6
|
-
LUIS FERNANDO LINHARES
-
**
-
Leader : ALISSON MARQUES DA SILVA
-
MEMBRES DE LA BANQUE :
-
ALISSON MARQUES DA SILVA
-
EDUARDO HABIB BECHELANE MAIA
-
GUSTAVO CAMPOS MENEZES
-
MAGALI REZENDE GOUVEA MEIRELES
-
Data: 4 avr. 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
Private transport has become a viable and increasingly popular option as an alternative to urban transport. With this growth, an old and recurrent problem emerges, that is the relationship between passenger demand and taxi supply. While taxi drivers spend considerable time looking for new passengers, on the other hand, some passengers cannot find taxi drivers. With this imbalance, taxi drivers pay more on fuel and depreciate the vehicle due to the excessive demand for passengers. Passengers end up not picking up taxis, causing disbelief in the availability of the service provided. This problem suggests the creation of techniques that make it possible to reduce the gap between the demand of passengers for taxis and the effective contingent of vehicles needed to meet this demand. In this context, this work presents a new approach to taxi demand prediction. The proposed approach uses historical data from taxi rides and weather data. The Kruskal-Wallis method identifies the most relevant variables, and an evolving fuzzy system is used for demand prediction. In this work, to perform the prediction, three evolving systems are evaluated: Autonomous Learning Multi-Model (ALMMo); Multivariable Gaussian Evolving Fuzzy Modeling System (eMG), and; evolving Neo-Fuzzy Neuron (eNFN). Computational experiments were conducted to compare the performance of the proposed approach. The results showed that the proposed approach achieved better or comparable results related to other published works, suggesting, therefore, promising applicability of this alternative as a solution to the problem of passenger supply and demand.
|
|
7
|
-
ARIANE HÉLIDA CRUZ LIMA
-
**
-
Leader : FLAVIO LUIS CARDEAL PADUA
-
MEMBRES DE LA BANQUE :
-
ADRIANO CESAR MACHADO PEREIRA
-
FLAVIO LUIS CARDEAL PADUA
-
MARCO AURÉLIO GUERRA PIMENTEL
-
MARCO TULIO ALVES NOLASCO RODRIGUES
-
NATALIA COSSE BATISTA
-
Data: 8 août 2022
-
Afficher le Résumé
-
**
|
|
8
|
-
FERNANDA SILVA COIMBRA
-
**
-
Leader : THIAGO MAGELA RODRIGUES DIAS
-
MEMBRES DE LA BANQUE :
-
RONALDO FERREIRA DE ARAÚJO
-
EMERSON DE SOUSA COSTA
-
GUSTAVO CAMPOS MENEZES
-
THIAGO MAGELA RODRIGUES DIAS
-
Data: 26 août 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
Scientific production has shown significant growth in recent times. Currently, there are numerous ways to propagate scientific knowledge. Therefore, scientific events emerge as one of the main means of disseminating research results in several areas of knowledge and aim to make information and scientific knowledge available to a group of people. And yet, publish the results of research in the format of annals of the events. This is a compilation of all the articles presented at a specific event. However, aiming at the integration of proceedings of different events, this task is not trivial, considering that each event has its publications in specific repositories and in different formats. In this context, integrating and analyzing the articles generated from the events becomes a complex task. The work in question aims to trace characteristics of scientific events and their published articles, using data extracted from the curricula registered on the Lattes Platform. After the collection of curricular data from the Lattes Platform in January 2021, steps of selection, treatment, curation of the data were carried out; featuring a set of 360,888 curricula of individuals with completed doctorates and 11,416,655 articles published in annals of events. In order to make the presented results possible, several analyzes were carried out in order to portray and detail the characteristics of scientific events, published articles and authors. The results presented will serve as a basis for several other studies that aim to understand the development of Brazilian science based on scientific events
|
|
9
|
-
DIEGO ASCANIO SANTOS
-
**
-
Leader : GUSTAVO CAMPOS MENEZES
-
MEMBRES DE LA BANQUE :
-
DANIEL HASAN DALIP
-
GUSTAVO CAMPOS MENEZES
-
LIVIA MARIA DUTRA
-
MICHELE AMARAL BRANDÃO
-
THIAGO MAGELA RODRIGUES DIAS
-
Data: 30 nov. 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
Chatbots with good quality are able to maintain service expectations and therefore, increase users’ retention, engagement, and satisfaction. Users’ opinion — available through surveys — is considered to measure chatbots’ quality, but, even if widely considered by academy and industry, it has several limitations such as: being subject to biases and careless evaluations from users, as well, the impossibility to being measured in real-time. To address these limitations, there are quality attributes that can mitigate them. Textual sentiment is one of those attributes, since textual sentiment is correlated to chatbots’ dialogs quality and can be measured in real-time. Therefore, through literature references that investigate the correlation between textual sentiment and chatbots’ quality, this work replicates correlation experiments over a specific dialogue corpora mentioned by one of the references, whose correlation results between sentiment and dialog quality weren’t shown, assuring that there is a correlation between textual sentiment and chatbots’ quality over this specific dialogue dataset.
|
|
10
|
-
GUILHERME VON RONDOW NASCIMENTO
-
**
-
Leader : BRENO RODRIGUES LAMAGHERE GALVAO
-
MEMBRES DE LA BANQUE :
-
BRENO RODRIGUES LAMAGHERE GALVAO
-
HEITOR AVELINO DE ABREU
-
THIAGO GOMES DE MATTOS
-
Data: 15 déc. 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
|
Thèses |
|
1
|
-
Daniel de Oliveira Capanema
-
**
-
Leader : ADRIANO CESAR MACHADO PEREIRA
-
MEMBRES DE LA BANQUE :
-
ADRIANO CESAR MACHADO PEREIRA
-
FLAVIO LUIS CARDEAL PADUA
-
ELIZABETH FIALHO WANNER
-
PAULO ROBERTO PEREIRA SANTIAGO
-
JOÃO GUSTAVO CLAUDINO
-
ADRIANO LIMA ALVES
-
Data: 27 févr. 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
2
|
-
AMALIA SOARES VIEIRA DE VASCONCELOS
-
**
-
Leader : RODRIGO TOMAS NOGUEIRA CARDOSO
-
MEMBRES DE LA BANQUE :
-
CLÁUDIA MAZZA DIAS
-
MARIA DA CONSOLAÇÃO MAGALHÃES CUNHA
-
ALLBENS ATMAN PICARDI FARIA
-
ELIZABETH FIALHO WANNER
-
JOSE LUIZ ACEBAL FERNANDES
-
RODRIGO TOMAS NOGUEIRA CARDOSO
-
Data: 24 mars 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
3
|
-
RENAN SANTOS MENDES
-
**
-
Leader : ELIZABETH FIALHO WANNER
-
MEMBRES DE LA BANQUE :
-
ALISSON MARQUES DA SILVA
-
CAROLINA GIL MARCELINO
-
ELIZABETH FIALHO WANNER
-
FLAVIO VINICIUS CRUZEIRO MARTINS
-
GUSTAVO CAMPOS MENEZES
-
IVAN REINALDO MENEGHINI
-
JOAO FERNANDO MACHRY SARUBBI
-
Data: 28 mars 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
In this work, a many objective formulation for the Vehicle Routing Problem with Demand Responsive Transport (VRPDRT) is addressed. The problem can be considered as a mode of transport similar to on-demand transportation services with shared races in which passengers are transported from their origin to their destination, sharing the same vehicle. The purpose of this type of transport is to reduce operating/displacement costs, meeting the passengers needs and offering high quality services. This work proposes the use of online cluster analysis based on Pearson’s and Kendall’s τ correlation coefficients to perform the dimensionality reduction at each generation of the MOEA/D algorithm, called OnCLτMOEA/D and OnCLρ-MOEA/D. The first sequence of experiments compared the following approaches: (i) offline clustering using Pearson’s correlation coefficient; (ii) offline clustering using Kendall’s τ correlation; (iii) online clustering using the Pearson correlation coefficient; (iv) online clustering using Kendall’s correlation coefficient τ; and (v) a baseline version, the MOEA/D in the original version. The algorithms were tested using a real dataset containing distances and travel times for Belo Horizonte city. To assess the dispersion of the solutions obtained by the algorithms, the hypervolume indicator was applied. The results of these experiments showed (i) there is no difference in the formulation obtained for the offline approaches; (ii) the online cluster-based algorithms are statistically better than the offline cluster-based algorithm; (iii) it is not possible to state that there is a statistical difference between the algorithms with online dimensionality reduction and the MOEA/D when the hypervolume indicator is used to evaluate the obtained solutions. Chord diagrams were applied to the obtained solutions and indicated that the diversity of solutions obtained by OnCLρ-MOEA/D and OnCLτ-MOEA/D are also slightly different. The second sequence of experiments performed compares the dimensionality reduction approaches based on online aggregation and another based on attribute selection. The approach using feature selection eliminates the objectives using two unsupervised feature selection techniques: Dispersion Rate (DR) and Maximum Variance (MV). Feature selection is used to select the cluster representative. The frequency of impact of dimensionality reduction on algorithm performance was also analyzed. Approaches using feature selection are coupled to the MOEA/D algorithm. Three algorithms were tested (i) online cluster using Pearson correlation coefficient with MOEA/D (OnCLρg-MOEA/D); (ii) online cluster using Pearson’s correlation coefficient and dispersion rate with MOEA/D (OnDRρg-MOEA/D); (iii) online cluster using Pearson’s coefficient and Maximum Variance with MOEA/D (OnMVρg-MOEA/D), where g = 1,25,50,100. The results of these new tests showed that, regardless of the frequency of the dimensionality reduction approach and the instances, the algorithms MOEA/D and OnCLρg-MOEA/D are statistically superior to OnDRρg-MOEA/D and OnMVρg-MOEA/D. It is not possible to state that there is a statistical difference between the results of the algorithms based on the selection of attributes online in terms of the hypervolume indicator. The chord diagrams showed that there is a greater diversity of solutions obtained when all objective functions are used in the reduced form of the problem and also when the algorithms remain for more generations in a given reduced formulation.
|
|
4
|
-
GUSTAVO ALVES FERNANDES
-
**
-
Leader : SERGIO RICARDO DE SOUZA
-
MEMBRES DE LA BANQUE :
-
HAROLDO GAMBINI SANTOS
-
ALEXANDRE XAVIER MARTINS
-
ELISANGELA MARTINS DE SA
-
GUSTAVO CAMPOS MENEZES
-
MAURÍCIO CARDOSO DE SOUZA
-
SERGIO RICARDO DE SOUZA
-
Data: 30 mars 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
5
|
-
RAFAELA PRISCILA CRUZ MOREIRA
-
**
-
Leader : ELIZABETH FIALHO WANNER
-
MEMBRES DE LA BANQUE :
-
HELENA SIQUEIRA VASSIMON
-
ANDRE RODRIGUES DA CRUZ
-
ELIZABETH FIALHO WANNER
-
FLAVIO VINICIUS CRUZEIRO MARTINS
-
GUSTAVO CAMPOS MENEZES
-
JOAO FERNANDO MACHRY SARUBBI
-
SERGIO RICARDO DE SOUZA
-
Data: 31 mars 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
This thesis addresses the Menu Planning Problem, presenting mathematical models to generate school menus that attending the requirements of the National Food Program School (PNAE). The two models have the function of minimizing the total cost of the menu. In the first model, we want to generate menus for 5 days and we have the following constraints: composition; color; consistency; variety; the minimum and the maximum limits of: carbohydrates, lipids and proteins, the maximum limit of sodium and saturated fat. In the second model, we want to generate menus for 푛 days and considered constraints related to: variety; the minimum and the maximum limits of: carbohydrates, proteins, lipids, sodium, saturated fat, and added sugar; minimum supply of: in natura or minimally processed foods, including weekly servings of fruits, greens and vegetables; weekly minimum offer of: vitamin A, foods with heme iron and sources of vitamin C when there are foods with non-heme iron; the maximum limit of: margarine, sweetened dairy products and meat derivatives products and finally, and combination and rejection of preparations. The menus were obtained by solving the linear mathematics models proposed through a mathematical programming software package performance (IBM CPLEX Optimizer). The qualitative aspects of the menus obtained were evaluated through the IQ COSAN - Safety Coordination Quality Index Nutritional Food. The results showed that the menus of both the first and second modeling are adequate to the IQ COSAN and to the recommendations of the PNAE. The costs were adequate, taking into account the financial resources made available to the schools. Cultural aspects of the menu were accounted for through the foods and culinary preparations included in the program. The results showed that the menus generated by the models are feasible and suitable both quantitatively and qualitatively for scholar menu planning. This study makes feasible and contributes to the pandemic scenario of obesity worldwide, presenting menus that introduce healthy eating habits, as well as helping the professional responsible for preparing them.
|
|
6
|
-
LARISSA LOPES LIMA
-
**
-
Leader : ALLBENS ATMAN PICARDI FARIA
-
MEMBRES DE LA BANQUE :
-
ROBERTO FERNANDES SILVA ANDRADE
-
MARIA HELENA FRANCO MORAIS
-
ALLBENS ATMAN PICARDI FARIA
-
ANDREA RODRIGUES MARQUES
-
ELIZABETH FIALHO WANNER
-
THIAGO GOMES DE MATTOS
-
Data: 13 mai 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
The field of study of complex systems is intertwined with many areas of science, being a promising approach to the study of biological problems. In this work, we applied some complex systems tools to study three biological problems of great current concern: the COVID-19 pandemic, the cyclicality of dengue outbreaks, and the impact of climate change on an invasive species. In the COVID-19 study, we developed an agent-based model to simulate the spread of the disease and the effect of superspreaders under different mobility scenarios. A network analysis was applied to investigate the role of superspreaders. The results showed that some individuals are crucial to the infection process. Furthermore, we performed simulations for some Brazilian cities using networks, mobility data extracted by cell phones, and different infection scenarios. In the case of the dengue study, we initially applied the visibility graph technique and entropy analysis to interpret two dengue time series data. Also, a tool called impact-frequency histogram was applied. This tool is generated from the visibility graph and showed better results for dengue cycling than the approach using entropy. Then, an agent-based model was adapted to investigate the dynamics of different scenarios of dengue. The result made it possible to identify humans and mosquitoes that play an essential role in the networks. Finally, we adapted an agent-based model to simulate the seed bank dynamics of an invasive species under different climate change scenarios. The results of the three models can help in decision-making in different areas. The COVID-19 and dengue models can be adapted for other scenarios and diseases. The invasive species model can also be adapted to assist in managing these species. Thus, the complex systems approach allowed us to achieve unprecedented results in treating and analyzing biologically motivated problems, such as identifying superspreaders in the spread of COVID-19, cyclicality in dengue epidemic outbreaks, and the influence of climate change on the seed bank of invasive species. All studies discussed in this work are likely to be continued, whether for the same problems or similar applications.
|
|
7
|
-
SABRINA MOREIRA DE OLIVEIRA
-
**
-
Leader : SERGIO RICARDO DE SOUZA
-
MEMBRES DE LA BANQUE :
-
THOMAS STÜTZLE
-
CAROLINA GIL MARCELINO
-
ELISANGELA MARTINS DE SA
-
ELIZABETH FIALHO WANNER
-
FLAVIO VINICIUS CRUZEIRO MARTINS
-
LEONARDO CÉSAR TEONÁCIO BEZERRA
-
SERGIO RICARDO DE SOUZA
-
Data: 24 juin 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
8
|
-
EMERSON EUSTÁQUIO COSTA
-
**
-
Leader : DOUGLAS ALEXANDRE GOMES VIEIRA
-
MEMBRES DE LA BANQUE :
-
LEONARDO WILLER DE OLIVEIRA
-
ADRIANO CHAVES LISBOA
-
DOUGLAS ALEXANDRE GOMES VIEIRA
-
ELISANGELA MARTINS DE SA
-
GUSTAVO RODRIGUES LACERDA SILVA
-
LUCAS SIRIMARCO MOREIRA GUEDES
-
RODNEY REZENDE SALDANHA
-
Data: 18 juil. 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
9
|
-
CARLA LILIANE GUEDES FONSECA
-
**
-
Leader : ARTHUR RODRIGO BOSCO DE MAGALHAES
-
MEMBRES DE LA BANQUE :
-
ARTHUR RODRIGO BOSCO DE MAGALHAES
-
ADRIANO CESAR MACHADO PEREIRA
-
ALLBENS ATMAN PICARDI FARIA
-
RODRIGO TOMAS NOGUEIRA CARDOSO
-
LUIS ALBERTO D AFONSECA
-
ADÉLCIO CARLOS DE OLIVEIRA
-
Data: 12 août 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
10
|
-
Fernando Garcia Diniz Campos Ferreira
-
**
-
Leader : RODRIGO TOMAS NOGUEIRA CARDOSO
-
MEMBRES DE LA BANQUE :
-
FREDERICO GADELHA GUIMARÃES
-
ADRIANO CESAR MACHADO PEREIRA
-
ARTHUR RODRIGO BOSCO DE MAGALHAES
-
FELIPE DIAS PAIVA
-
RODRIGO TOMAS NOGUEIRA CARDOSO
-
Data: 24 nov. 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
11
|
-
CLAUDIO LUCIO DO VAL LOPES
-
MULTI- AND MANY-OBJECTIVE OPTIMIZATION: SOME ADVANCES TOWARDS THEORETICAL ASPECTS IN PERFORMANCE QUALITY INDICATORS AND EVOLUTIONARY FRAMEWORKS
-
Leader : FLAVIO VINICIUS CRUZEIRO MARTINS
-
MEMBRES DE LA BANQUE :
-
FLAVIO VINICIUS CRUZEIRO MARTINS
-
ADRIANO CHAVES LISBOA
-
ELISANGELA MARTINS DE SA
-
ELIZABETH FIALHO WANNER
-
RICARDO HIROSHI CALDEIRA TAKAHASHI
-
CARLOS MANUEL MIRA DA FONSECA
-
KALYANMOY DEB
-
Data: 30 nov. 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
Many-Objective Optimization (MaO) refers to optimization problems having four or more objectives, the increase in objective dimensionality brings some complex issues, such as the ineffectiveness of the Pareto dominance relation, quality indicators calculation, solution sets visualization, balancing convergence and diversity, and others. A key issue in many- and multi-objective optimization is comparing and assessing solution sets obtained by optimization algorithms. This is not a simple task; the outcome of many-objective optimization algorithms is typically a set of incomparable solutions. Using a quality indicator to reflect the inner Pareto front characteristics requires careful design/selection of such indicators. In the first part of this thesis, we deal with the Dominance move (DoM) quality indicator. We propose novel approaches to calculate DoM using mixed-integer programming (MIP) models and an approximate method using machine learning techniques. In general, our attempts uses the dominance move quality indicator as a suitable way to measure, compare, and assess many-objective problems. Another challenge in MaO is to provide a true representative set with the desired number of Pareto-optimal solutions in a reliably well-distributed set. In the second part of this work, we propose a multi-stage framework involving reference-vector-based evolutionary multi- and many-objective algorithms that attempt to rectify previous stages’ shortcomings by careful executions of subsequent stages so that a prescribed number of well-distributed and well-converged solutions are achieved. The results presented in this thesis come from the attempts to address challenges in evolutionary many- and multi-objective optimization. This research has not only analyzed and systematically evaluated the existing approaches but has also extended them in innovative directions related to quality indicators and improvements in multistage approach for convergence and diversity balance in evolutionary algorithms.
|
|
12
|
-
ALEXANDRE FRIAS FARIA
-
**
-
Leader : SERGIO RICARDO DE SOUZA
-
MEMBRES DE LA BANQUE :
-
CARLOS ALEXANDRE SILVA
-
CAROLINA GIL MARCELINO
-
EDUARDO UCHOA BARBOZA
-
ELISANGELA MARTINS DE SA
-
GUSTAVO CAMPOS MENEZES
-
MARCONE JAMILSON FREITAS SOUZA
-
SERGIO RICARDO DE SOUZA
-
Data: 16 déc. 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
13
|
-
ANDERSON MOREIRA DE VASCONCELOS
-
**
-
Leader : MARCONE JAMILSON FREITAS SOUZA
-
MEMBRES DE LA BANQUE :
-
ELISANGELA MARTINS DE SA
-
ELIZABETH FIALHO WANNER
-
GUSTAVO CAMPOS MENEZES
-
MARCONE JAMILSON FREITAS SOUZA
-
MARIA AMÉLIA LOPES SILVA
-
MATHEUS NOHRA HADDAD
-
SERGIO RICARDO DE SOUZA
-
Data: 20 déc. 2022
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|