|
Thèses |
|
1
|
-
DENISE LEMOS CARDOSO
-
**
-
Leader : ALLBENS ATMAN PICARDI FARIA
-
MEMBRES DE LA BANQUE :
-
ALLBENS ATMAN PICARDI FARIA
-
BRENO RODRIGUES LAMAGHERE GALVAO
-
FELIPE GALVÃO RAFAEL MAGALHÃES
-
Data: 10 févr. 2023
-
Afficher le Résumé
-
**
|
|
2
|
-
CESAR AUGUSTO SOUZA DE OLIVEIRA
-
**
-
Leader : SERGIO RICARDO DE SOUZA
-
MEMBRES DE LA BANQUE :
-
SERGIO RICARDO DE SOUZA
-
GUSTAVO CAMPOS MENEZES
-
MARIA AMÉLIA LOPES SILVA
-
Rodney Oliveira Marinho Diana
-
Data: 24 févr. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
3
|
-
DIOGO SOUZA DE FIGUEIREDO
-
**
-
Leader : THIAGO DE SOUZA RODRIGUES
-
MEMBRES DE LA BANQUE :
-
GUSTAVO CAMPOS MENEZES
-
SANDRO RENATO DIAS
-
THIAGO DE SOUZA RODRIGUES
-
Data: 27 févr. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
In almost every daily activity, the internet is always present. Many of these activities require the sharing of personal data (e.g. address, telephone, age, location, among others) and data considered sensitive (e.g. medical diagnosis, bank account, among others), generating a privacy concern. of this information to society. Sharing information requires the use of data protection tools, so the information in this data cannot be used to identify an individual. In order to comply with applicable laws and regulations, as well as informed data protection policies, a data protection mechanism called data anonymization has emerged. This mechanism basically consists of removing identifiers and hiding sensitive data, making it impossible to re-identify. As large amounts of data from individuals are disseminated, new challenges arise for the protection of their privacy. Various anonymization algorithms have been proposed, making publishing privacy-preserving data a very broad area of research. However, it is difficult to identify and select the most suitable algorithm, due to the large number of available algorithms and limited information about their performance. In this work, three anonymization algorithms called DataFly, Incognito and Mondrian are presented. Its efficiencies in data processing, its effectiveness in the amount of data used will be addressed, employing a wide set of different parameters, metrics and data sets. To guide the selection of an algorithm, a battery of experiments will be carried out among them to identify which factors can influence performance, presenting the conditions in which each algorithm outperforms each other in certain privacy requirements as well as their advantages and disadvantages
|
|
4
|
-
ALEXANDRE AUGUSTO AGUIAR GOMES
-
**
-
Leader : FLAVIO LUIS CARDEAL PADUA
-
MEMBRES DE LA BANQUE :
-
DANIEL HASAN DALIP
-
FLAVIO LUIS CARDEAL PADUA
-
GUSTAVO CAMPOS MENEZES
-
MICHELE AMARAL BRANDÃO
-
Data: 27 févr. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
5
|
-
WANDERLEY DA SILVA
-
**
-
Leader : ANTONIO PAULO BAETA SCARPELLI
-
MEMBRES DE LA BANQUE :
-
OSVALDO NEGRINI NETO
-
ANTONIO PAULO BAETA SCARPELLI
-
GIANCARLO QUEIROZ PELLEGRINO
-
LUIZ CLEBER TAVARES DE BRITO
-
Data: 13 mars 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
6
|
-
RODRIGO COIMBRA PEREIRA ALVES
-
**
-
Leader : ARTHUR RODRIGO BOSCO DE MAGALHAES
-
MEMBRES DE LA BANQUE :
-
ALLBENS ATMAN PICARDI FARIA
-
ARTHUR RODRIGO BOSCO DE MAGALHAES
-
CHARLENE CASSIA DE RESENDE
-
CRISTIANO ARBEX VALLE
-
Data: 4 juil. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
7
|
-
ROMARIO DOS SANTOS LOPES DE ASSIS
-
**
-
Leader : MARCONE JAMILSON FREITAS SOUZA
-
MEMBRES DE LA BANQUE :
-
TATIANA ALVES COSTA
-
EDUARDO CAMARGO DE SIQUEIRA
-
ELISANGELA MARTINS DE SA
-
MARCONE JAMILSON FREITAS SOUZA
-
MARIA AMÉLIA LOPES SILVA
-
SERGIO RICARDO DE SOUZA
-
Data: 11 juil. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
8
|
-
LEONARDO CORRÊA CARDOSO
-
**
-
Leader : ELISANGELA MARTINS DE SA
-
MEMBRES DE LA BANQUE :
-
ELISANGELA MARTINS DE SA
-
GUSTAVO CAMPOS MENEZES
-
Rodney Oliveira Marinho Diana
-
SERGIO RICARDO DE SOUZA
-
Data: 14 juil. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
9
|
-
RÓGENES REIS FIRMINO
-
**
-
Leader : FLAVIO LUIS CARDEAL PADUA
-
MEMBRES DE LA BANQUE :
-
FABÍOLA DE OLIVEIRA PAES LEME
-
FLAVIO LUIS CARDEAL PADUA
-
GUSTAVO CAMPOS MENEZES
-
MARCO TULIO ALVES NOLASCO RODRIGUES
-
NATALIA COSSE BATISTA
-
Data: 7 août 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
10
|
-
JOICE MARTINHA RODRIGUES
-
**
-
Leader : GUSTAVO CAMPOS MENEZES
-
MEMBRES DE LA BANQUE :
-
ELISANGELA MARTINS DE SA
-
GUSTAVO ALVES FERNANDES
-
GUSTAVO CAMPOS MENEZES
-
THIAGO DE SOUZA RODRIGUES
-
Data: 5 oct. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
11
|
-
MIRIAN GREINER DE OLIVEIRA PINHEIRO
-
**
-
Leader : GRAY FARIAS MOITA
-
MEMBRES DE LA BANQUE :
-
ELISANGELA MARTINS DE SA
-
GRAY FARIAS MOITA
-
GUILHERME DE CASTRO LEIVA
-
MATHEUS HENRIQUE DE SOUSA OLIVEIRA
-
RENATO GUIMARAES RIBEIRO
-
Data: 23 oct. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
12
|
-
DANIEL LETO NEIVA COSTA
-
**
-
Leader : ELIZABETH FIALHO WANNER
-
MEMBRES DE LA BANQUE :
-
ELIZABETH FIALHO WANNER
-
FLAVIO VINICIUS CRUZEIRO MARTINS
-
ARTHUR RODRIGO BOSCO DE MAGALHAES
-
JUAN PABLO NEIROTTI
-
ROBERTO CASTRO ALAMINO
-
Data: 18 déc. 2023
-
Afficher le Résumé
-
**
|
|
13
|
-
GERMANO TEIXEIRA DE MIRANDA
-
**
-
Leader : FLAVIO VINICIUS CRUZEIRO MARTINS
-
MEMBRES DE LA BANQUE :
-
LENIANA SANTOS NEVES
-
ALISSON MARQUES DA SILVA
-
FLAVIO VINICIUS CRUZEIRO MARTINS
-
Data: 21 déc. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
Computer diagnostic aid systems are used in the various health sectors to facilitate the work of the most used health professionals. These computer systems help healthcare professionals to detect the chance of error, correct the accuracy of the diagnosis, in addition to reducing the time and cost of making the diagnosis. More specifically in the measures traced from cephalometric treatment in linear and not crucial to determine an orthodontic treatment. For the elaboration of this exam, measurements are made in a lateral medicine of the head called teleradiography. The diagnosis of a tracing tracing was obtained through the evaluation of measurements performed on teleradiography in comparison with the defined ideal pattern. This can be a comprehensive and well-defined feature as a candidate for the application of computational assessment techniques, without the possibility of automatic diagnosis, an intervention manual of the specialist. Research is evaluated as an existing computational research technique and currently applied. For the application of computer vision techniques, we intend to extract a basis from one of cephalometric tracing training tests performed in the most computerized tracing data collection tool of Cfaznet, one of the tracing data software in Latin America. . This training base will be used for nutritional calculation as a computer vision technique that can best perform the task of anatomical points in a teleradiography. In order to find the techniques it is best to carry out the tracing exam, most likely the job of searching to find the best techniques comparatively. In this way, computer vision techniques were studied today capable of performing a projected cephalometric determination exam without a theoretical intervention.
|
|
|
Thèses |
|
1
|
-
SOLANGE CONCEIÇÃO DE OLIVEIRA
-
**
-
Leader : LEONARDO DOS SANTOS LIMA
-
MEMBRES DE LA BANQUE :
-
ADRIANO CESAR MACHADO PEREIRA
-
ADÉLCIO CARLOS DE OLIVEIRA
-
FELIPE DIAS PAIVA
-
JOSE LUIZ ACEBAL FERNANDES
-
LEONARDO DOS SANTOS LIMA
-
Data: 23 févr. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
2
|
-
PRISCILA FLÁVIA SOUZA DA SILVA
-
**
-
Leader : GRAY FARIAS MOITA
-
MEMBRES DE LA BANQUE :
-
ALISSON MARQUES DA SILVA
-
ELIENE PIRES CARVALHO
-
GRAY FARIAS MOITA
-
GUSTAVO CAMPOS MENEZES
-
MOACIR KRIPKA
-
ROQUE LUIZ DA SILVA PITANGUEIRA
-
Data: 24 févr. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
3
|
-
JOSENILDO SILVA DE LIMA
-
**
-
Leader : RODRIGO TOMAS NOGUEIRA CARDOSO
-
MEMBRES DE LA BANQUE :
-
ANA PAULA PINTADO WYSE
-
ARTHUR RODRIGO BOSCO DE MAGALHAES
-
JOSE LUIZ ACEBAL FERNANDES
-
RICARDO HIROSHI CALDEIRA TAKAHASHI
-
RODRIGO TOMAS NOGUEIRA CARDOSO
-
Data: 27 févr. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
4
|
-
SINAIDE NUNES BEZERRA
-
**
-
Leader : MARCONE JAMILSON FREITAS SOUZA
-
MEMBRES DE LA BANQUE :
-
ELISANGELA MARTINS DE SA
-
ELIZABETH FIALHO WANNER
-
LUIZ SATORU OCHI
-
MARCONE JAMILSON FREITAS SOUZA
-
PUCA HUACHI VAZ PENNA
-
SERGIO RICARDO DE SOUZA
-
Data: 27 févr. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
5
|
-
EMMANUEL TAVARES FERREIRA AFFONSO
-
**
-
Leader : ALISSON MARQUES DA SILVA
-
MEMBRES DE LA BANQUE :
-
ALISSON MARQUES DA SILVA
-
DANIEL FURTADO LEITE
-
ELISANGELA MARTINS DE SA
-
ELIZABETH FIALHO WANNER
-
GRAY FARIAS MOITA
-
MAURÍLIO JOSÉ INÁCIO
-
Data: 28 févr. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
This work aims to explore different areas and solutions to solve classification and pattern recognition problems, proposing a new method for feature selection (FS) and also a new evolving fuzzy clustering model for data classification. The proposed FS approach, called Mean Ratio for Feature Selection (MRFS), aims to improve the ability of classification algorithms with only low computational cost equations and with basic mathematical operations such as addition, division and comparison. MRFS uses the ratio of the averages associated with the different outputs of the data sets to be classified, and the value of the ratio between the averages is used to select the most relevant attributes for a classifier. The proposed classification model, called the evolving Fuzzy Mean Classifier (eFMC), is a fuzzy clustering model of data that has incremental learning characteristics, capable of processing continuous flow data, extracting knowledge when new samples are presented, identifying changes in the standard of the data and maintain the quality of the response with its processing. The experiments with both proposals were carried out as follows: MRFS in a first instance was performed as a filter comparing with other existing FS methods; in a second instance, MRFS was added as a wrapper in different classifiers and its results were compared with the same classifiers without MRFS; the experiments with the eFMC had their performance and accuracy compared with other evolving models, being executed in different databases. The experiments of both proposals were promising, with good accuracy in the final results. In this way, the current phase of this work lies on a FS solution and on a new evolving algorithm for data processing in classification problems. For this last solution, the development of a new evolving model with the ability to select input attributes during processing is estimated, increasing its predictive capability without a large increase in computational cost.
|
|
6
|
-
FLÁVIA DA SILVA CORDEIRO
-
**
-
Leader : JOSE LUIZ ACEBAL FERNANDES
-
MEMBRES DE LA BANQUE :
-
ALLBENS ATMAN PICARDI FARIA
-
ARTHUR RODRIGO BOSCO DE MAGALHAES
-
CLÁUDIA MAZZA DIAS
-
FABIO ROCHA DA SILVA
-
JOSE LUIZ ACEBAL FERNANDES
-
LUCY TIEMI TAKAHASHI
-
Data: 28 févr. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
7
|
-
LUIZ OTAVIO RODRIGUES ALVES SERENO
-
**
-
Leader : JOSE LUIZ ACEBAL FERNANDES
-
MEMBRES DE LA BANQUE :
-
JOSE LUIZ ACEBAL FERNANDES
-
ARTHUR RODRIGO BOSCO DE MAGALHAES
-
JOSE GERALDO PEIXOTO DE FARIA
-
LUIS ALBERTO D AFONSECA
-
Ana Paula de Paiva Pereira
-
CARLOS ALEXANDRE SILVA
-
Data: 28 févr. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
TNormal diffusion is a process observed in many phenomena in nature. For these processes there are well-developed simulations and theory, namely the random walk (RW) and the diffusion equation, or Fokker-Planck. However, in systems with long-range interactions between individuals or between individuals and the environment, there may be deviations from this regime and present anomalous behaviour. This class of more general processes is called anomalous diffusion. Systems in which there are interactions between individuals or between individuals and the environment, for example, deviate from the usual description of diffusion, presenting an anomalous behaviour. The theoretical approach to these phenomena ranges from analytical representation through partial differential equations, stochastic equations and computational simulations that mimic the phenomenon. From displacement data and characteristic times of a phenomenon on a microscopic scale, it is possible to identify the theoretical parameters of the models that macroscopically describe these data. For this purpose, a theoretical study is carried out in which datasets are generated by simulations with controlled microdynamic parameters and adjusted to a function associated with the phenomenon. In this study, flights are generated via continuous time random walk (CTRW) and the anomalous diffusion models are fractional partial differential equations with solutions obtained by Fractional Calculus. An evolutionary algorithm capable of obtaining parameters that describe the phenomenon is presented, with only a limitation in the parameter space being a prerequisite. Furthermore, in the discretization process of the theoretical model associated with the phenomenon, new parameters emerge whose behaviours are investigated, as their role can influence the study of anomalous diffusion. As an immediate application of superdiffusion is included the study of specialized proteins, called transcription factors, in cells that allow the transcription of deoxyribonucleic acid (DNA).
|
|
8
|
-
IZABELA MARQUES DE OLIVEIRA
-
**
-
Leader : ALLBENS ATMAN PICARDI FARIA
-
MEMBRES DE LA BANQUE :
-
ALLBENS ATMAN PICARDI FARIA
-
ANTONIO PAULO BAETA SCARPELLI
-
ARTHUR RODRIGO BOSCO DE MAGALHAES
-
BRUNA AMIN GONÇALVES
-
LAURA CORINA CARPI
-
TIAGO ALVES SCHIEBER FARIA
-
Data: 20 mars 2023
-
Afficher le Résumé
-
Air transportation remains the best option to connect distant regions and support world social and economic development, despite environmental challenges. Brazil depends strongly in this mode of transportation due to its continental dimensions. In this work, we conducted a study using the multiplex approach of complex networks to analyze the air transport network of Brazilian passengers, BATMN. In this context, we create the multiplex efficiency index (MEI) and its versions for targeted and/or weighted networks, which allow to quantify the level of connection heterogeneity of a multiplex network on a scale of 0 to 1, considering the characteristics of direction and importance level of edges, being appropriate for research in transportation networks. We used the MEI to evaluate the diversification of routes of airlines in Brazil between 2010 and 2021, a period of occurrence of major events in the country. We analyzed the influence of some of these events on network efficiency, such as the FIFA World Cup in 2014 and the Rio Olympic Games in 2016, as well as other sector changes, including the merger between airlines, airport privatization, the closure of operations of an important air carrier and COVID-19 pandemic. In addition, we investigated whether the diversification of routes affected the prices of air tickets in Brazil. Additionally, through a recently introduced methodology in the literature, we examined the network’s diffusion capacity and observed the relative gain that occurred in the diffusion of the air transportation system when the flight connections between different airlines are considered, outlining a scenario possible partnerships between airlines. The evaluations carried out in this work provide new and significant information about the passenger air transportation network in Brazil, contributing to optimize the offer of this type of transportation in the country and, consequently, reduce the costs involved. In addition, the proposed methodology allows not only a differentiated analysis of air transportation networks, but also the quantification of the efficiency of multiplex networks in general.
|
|
9
|
-
OTAVIANO MARTINS MONTEIRO
-
**
-
Leader : THIAGO DE SOUZA RODRIGUES
-
MEMBRES DE LA BANQUE :
-
THIAGO DE SOUZA RODRIGUES
-
ALISSON MARQUES DA SILVA
-
GUSTAVO CAMPOS MENEZES
-
ANDRE RODRIGUES DA CRUZ
-
ROGERIO MARTINS GOMES
-
SANDRO RENATO DIAS
-
Data: 24 août 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
10
|
-
GILSON VIEIRA SOARES
-
**
-
Leader : JOSE GERALDO PEIXOTO DE FARIA
-
MEMBRES DE LA BANQUE :
-
ADÉLCIO CARLOS DE OLIVEIRA
-
ALLBENS ATMAN PICARDI FARIA
-
GIANCARLO QUEIROZ PELLEGRINO
-
JOSE GERALDO PEIXOTO DE FARIA
-
LUIS ARGEL POVEDA CALVINO
-
MAURICIO REIS E SILVA JUNIOR
-
THIAGO GOMES DE MATTOS
-
Data: 29 août 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
In this study, we investigate the two-mode Bose-Hubbard model with atom-atom interactions harmonically tuned over time. The classical Hamiltonian is derived through an expansion of the quantum Hamiltonian in terms of 1/N. Chaotic behavior is observed in classical dynamics, and we map the parameter and phase space using the dominant Lyapunov exponent and quantum measures such as Loschmidt echo, generalized purity, and linear entropy. In the quantum counterpart of chaos, irregular Josephson oscillations are noted. In the regular regime, we identify collapses, revivals, quasi-periodicity, and coherent control of Josephson oscillations.
|
|
11
|
-
Túlio Philipe Ferreira e Vieira
-
**
-
Leader : GRAY FARIAS MOITA
-
MEMBRES DE LA BANQUE :
-
MARCOS PAULINO RORIZ JUNIOR
-
ALISSON MARQUES DA SILVA
-
ELISANGELA MARTINS DE SA
-
GRAY FARIAS MOITA
-
GUILHERME DE CASTRO LEIVA
-
MAGALI REZENDE GOUVEA MEIRELES
-
RENATO GUIMARAES RIBEIRO
-
Data: 29 août 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
According to data from the United Nations (UN), there are currently more than 3.8 billion people living in cities and in 2050, almost 70% of the world’s population will live in urbana reas. Cities are considered places of economic, political, cultural and social development, but disorderly urbanization, especially in developing countries, has deteriorated the quality of various systems and services offered to citizens. Among the systems that make up the urban infrastructure, the transportation system can be seen as an indicator of the residents’ quality of life. This is because, through the analysis of the transport system, from wich people gain access to medical, educational, cultural and leisure services, it is possible to identify social, economical and cultural differences. Thus, in order to ensure the construction of sustainable cities that offer a good quality of life to their inhabitants, the planning and efficient management of urban mobility is essential. In order to contribute to the construction of an efficient transport system, this work presents the development of a framework composed of low cost hardware and software, which can be used to generate and monitor real time indicators for the transport system. The proposedframeworkuses data from different sources such asprobe request, census and smart card and employs machine learning algorithms such as Kalman Filter and Artificial Neural Networks to generate the origin-destination matrix and other indicators of the transportation system. Preliminary results indicate that the strategy used manages to reduce errors in forecasting indicators to less than 5%.
|
|
12
|
-
CRISTIANO FRAGA GUIMARÃES NUNES
-
**
-
Leader : FLAVIO LUIS CARDEAL PADUA
-
MEMBRES DE LA BANQUE :
-
ADRIANO CESAR MACHADO PEREIRA
-
Celso Luiz de Souza
-
DARLAN NUNES DE BRITO
-
ELIZABETH FIALHO WANNER
-
FLAVIO LUIS CARDEAL PADUA
-
Data: 31 août 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
**
|
|
13
|
-
LUCIENE LOPES BORGES MIRANDA
-
**
-
Leader : LEONARDO DOS SANTOS LIMA
-
MEMBRES DE LA BANQUE :
-
GIANCARLO QUEIROZ PELLEGRINO
-
THIAGO DE SOUZA RODRIGUES
-
THIAGO GOMES DE MATTOS
-
LUIS ARGEL POVEDA CALVINO
-
LEONARDO FERREIRA CALAZANS
-
Data: 30 nov. 2023
-
Afficher le Résumé
-
The complexity of behavior and studies on the financial market and the Itô-Langevin Calculations, become an attractive field of studies for researchers in several areas, such as: Economics, Systems Analysis, Mathematics and Physics. We can emphasize its importance for such studies in interdisciplinary areas, such as: Econophysics, which relate elements of Economics and Physics in the study of markets and financial systems. Statistical mechanics is one of the main tools for modeling these financial systems, more precisely the price dynamics. To understand such facts, the Itô stochastic differential equations were used, with additive white noise as a mathematical model for the price dynamics of the financial market. Mathematical methods of Statistical Physics have been applied in the description of economic systems to time series to investigate the behavior of financial markets at different scales. Among the tools used for this, are measures of central dispersion, probability distribution, stochastic dynamics, and methods to estimate the correlations. Financial instruments such as assets, options and stock exchange indices fluctuate over time, so they can be modeled by stochastic processes. In this way this thesis intends to verify if a stochastic differential equation (interfered by white and non-white noise) attend a stylized fact of the market. More specifically, if the heavy tail of the distribution of absolute returns generated by the proposed equation follows an empirical power law. Furthermore, we estimate the Hurst exponent through R/S Analysis and DFA to verify the long-range dependence, or long memory of the obtained time series. As the noise term that composes the equation is non-differentiable, we introduce the stochastic calculus to analytically obtain its solution.
|
|
14
|
-
BRUNO ZANOTELLI FELIPPE
-
**
-
Leader : ANTONIO PAULO BAETA SCARPELLI
-
MEMBRES DE LA BANQUE :
-
MARCOS DONIZETE RODRIGUES SAMPAIO
-
ALEXANDRE RODRIGUES VIEIRA
-
ANTONIO PAULO BAETA SCARPELLI
-
ARTHUR RODRIGO BOSCO DE MAGALHAES
-
JOSE LUIZ ACEBAL FERNANDES
-
LUIS ALBERTO D AFONSECA
-
Data: 4 déc. 2023
-
Voir Mémoire/Thèse
-
Afficher le Résumé
-
The perturbative calculations in Quantum Field Theory can be represented, graphically, in terms of Feynman diagrams. The calculations of amplitudes represented by Feynman diagrams involve integrals, often divergent beyond tree level. These integrals require a treatment known as regularization, in order to extract the physical content of the amplitude. One of the most powerful techniques, Dimensional Regularization, involve dimensional extensions of space-time. However, there are models that involve mathematical objects whose the dimensional extension is ambiguous or ill-defined. A technique, called Implicit Regularization, proved to be appropriate for the treatment of models whose dimensional extension is not well defined. The method allows the calculations to be performed in the own theory dimension and prescribes simple rules for the preservation of the symmetries of the models. However, for a single Feynman integral, the expansion to separate divergences can generate a large set of finite integrals in the case of high degrees of divergence. In addition, expansions generate high powers of momenta in the numerator and denominator, which results in long and labor-intensive calculations. In this thesis, we developed a new procedure for the application of Constrained Implicit Regularization that simplifies the calculation of amplitudes, including finite parts.
|
|