|
Dissertations |
|
1
|
-
CLEIDIANA REIS DOS SANTOS
-
Desafio IoT: Serious Game for Immersion in Embedded Software Development in the Context of Smart Homes
-
Advisor : RODRIGO DUARTE SEABRA
-
COMMITTEE MEMBERS :
-
BRUNO GUAZZELLI BATISTA
-
LUIS HENRIQUE NUNES
-
RODRIGO DUARTE SEABRA
-
Data: Feb 13, 2023
-
-
Show Abstract
-
The technological evolution provided by the Internet of Things is in increasing development and the demand for professionals increases proportionally. Parallel to this reality, the use of games as a learning tool is a suitable practice especially for the younger audience, as they present elements of fun and engagement. Serious games can help players acquire new experiences and complex knowledge, which are obtained by solving the challenges. In this context, the serious game proposed in this research, Desafio IoT, aims to provide an overview of some problems and solutions in embedded software development for smart homes. In addition to a serious purpose, the game seeks to awaken students' interest in the subject, spreading the idea and motivating to work in the development area. The implementation of the game was carried out according to the Learning Mechanics – Game Mechanics (LM-GM) specification model. In order to investigate the educational impact provided by the experience of using the game, besides the questionnaires on usage and technical knowledge, the MEEGA+ questionnaire was used to evaluate the game. The results allow concluding that the game was able to introduce students to the Internet of Things area and motivate them to further their knowledge on the subject. The evaluation of the game by the students presented a positive overall result, as well as approval in seven of the eight dimensions used in the analysis.
|
|
2
|
-
JONAS HENRIQUE RIBEIRO PAULA
-
Collaborative Committee: The use of a collaborative system in the development of participatory legal instruments
-
Advisor : MELISE MARIA VEIGA DE PAULA
-
COMMITTEE MEMBERS :
-
JONICE OLIVEIRA SAMPAIO
-
ADLER DINIZ DE SOUZA
-
MELISE MARIA VEIGA DE PAULA
-
Data: Feb 24, 2023
-
-
Show Abstract
-
The Cities Statute (CS) regulates articles 182 and 183 of the Federal Constitution, establishing a set of participatory legal instruments that combined determine how urban policy should be conducted throughout the country. However, the application of the CS represents a challenge for the municipal administration, one of the reasons is the demand for popular participation in the management of urban policy to ensure the equity of citizens' rights. To promote this participation representative committees are created with members who represent the parts of society in the municipality. However, the members of the committees involved in the elaboration of the CS instruments, in most cases, have divergent demands on public choices, conflicting agendas and other characteristics that make it difficult for the work to be conducted in a harmonious and efficient way. Therefore, it is always necessary to try to encourage collaboration so that everyone involved can present their point of view and actively participate. In this way, this work proposes the analysis of the intervention of a collaborative system, called Collaborative Committee, in the activities conducted by committees during the elaboration of participatory legal instruments of the CS. The Collaborative Committee was implemented in real projects conducted by a research and extension group from the Federal University of Itajubá that helps municipalities in the elaboration and revision of these instruments, following the guidelines of the action research methodology. In the analysis of the results, it was possible to notice that there was an improvement in the coordination of activities, an advance in the cooperation between those involved and better opportunities for communication between the members with the intervention of the Collaborative Committee.
|
|
3
|
-
IGOR MOREIRA ALVES
-
Ideb, the federation units and the profile of public schools: an unsupervised exploratory data analysis.
-
Advisor : CARLOS HENRIQUE DA SILVEIRA
-
COMMITTEE MEMBERS :
-
CARLOS HENRIQUE DA SILVEIRA
-
MELISE MARIA VEIGA DE PAULA
-
CRISTIANE NERI NOBRE
-
Data: Feb 24, 2023
-
-
Show Abstract
-
This dissertation seeks to use data science, through unsupervised exploratory analysis, to prove what the literature already knows and/or present new discoveries about the influence of infrastructure on Brazilian basic education. School infrastructure is not limited to the architectural issue of schools, but also to the educational and administrative environment, equipment, educational resources, practices, curricula and the teaching and learning process. Data collection was carried out on open data from the 2019 School Census (Basic Education) and the Basic Education Development Index (Ideb), for the years 2005-2019. The choice of the year 2019 was because it was the last year that schools presented results before the influence of the COVID-19 pandemic. After several data treatments and the choice of attending only the initial segment of fundamental education, two analysis methodologies were applied: Correlogram and Factor Analysis (FA). For clarity in the results, new attributes were created referring to the federative entities that allowed identifying which states and school profiles are better related to the growth and good results of the Ideb. For these correlations, the Sigma of the Gaussian Copula was chosen, which takes into account the categorical and continuous data and also generated a definite positive matrix. The Correlogram generated a square matrix that presented the attribute relationships in a Heatmap Dendrogram. Divided into 4 large groups, each one had specific characteristics and relationships with federal entities. The first group had a strong relationship with basic infrastructure; the second group, with IDEB and the most sophisticated infrastructures; the third group showed few relationships between the attributes; and the last group had strong negative correlations and contained greater precariousness in infrastructure. After verifying the compatibility of the database for the application of the FA, it was estimated that 10 factors would be suitable for this study. Four factors were associated with the attributes of the Ideb, the focus of this work. Three patterns were also observed in the attributes that listed good results in the Ideb with different infrastructures, policies and/or educational proposals: the first group, guided by São Paulo state, presented basic sanitation offered by the public service, quality internet for use in learning and institutions schoolchildren; the second group, headed by Minas Gerais state, indicates an association with flexibility in traditional teaching, with school cycles and non-serial classrooms; the third group was marked by complementary activities and specialized care, represented by the Ceará state . In contradiction to these parameters, schools with the EJA modality, mainly in the northeast, tend to have lower results in the Ideb. The other 6 factors added a lot of relevant information, including those related to the correlations and anti-correlations of the federative entities and specific attributes. As seen, data science has a lot to add to the field of education. Future works are expected to add even more data, such as longitudinal studies on the Ideb and to add other educational indices such as the Ioeb and the socioeconomic level of the population.
|
|
4
|
-
DOMINGOS SAVIO FARIA PAES
-
Predictive detection of anomalies in computer networks using machine learning.
-
Advisor : BRUNO GUAZZELLI BATISTA
-
COMMITTEE MEMBERS :
-
BRUNO GUAZZELLI BATISTA
-
CARLOS HENRIQUE VALERIO DE MORAES
-
LOURENCO ALVES PEREIRA JUNIOR
-
RAFAEL DE MAGALHAES DIAS FRINHANI
-
Data: Feb 27, 2023
-
-
Show Abstract
-
With the increasing dependence on technologies on a daily basis, it is evident the con- cern to maintain the infrastructures that support its operation, thus guaranteeing a good experience for the end user. Thus, denial of service attacks are among the main causes of anomalies in computer networks, which can cause degradation or even interruption of services. In this context, the application of new technologies, such as artificial intelligence or machine learning, becomes increasingly necessary to ensure more agility in detecting problems, reducing their impacts. Thus, this work presents an analysis between different methods of classifier supervised machine learning, applied to data collected fromnetwork equipment, of the switch type, in order to detect anomalies in the network infrastruc- ture of a higher education institution. The machine learning methods used in this work were: Decision Tree, Random Forest, Extra Tree, Gradient Boosting, Extreme Gradient Boosting and Histogram Gradient Boosting. The models generated from these methods showed promise, being able to achieve results with 99.88% in the Weighted F1 metric and 99.16% of Balanced Accuracy. Other points, such as training time, prediction time and save file size, were also taken into account for the classification of the best method. Given the importance of fault detection tools, this work contributes to the definition of the best approaches and thus allows the development of new and more efficient tools for this purpose.
|
|
5
|
-
Otávio Soares Silva
-
Differential Transconductance Amplifier Characterization Based on CMOS Inverters
-
Advisor : RODRIGO APARECIDO DA SILVA BRAGA
-
COMMITTEE MEMBERS :
-
DALTON MARTINI COLOMBO
-
PAULO MARCOS PINTO
-
RODRIGO APARECIDO DA SILVA BRAGA
-
SANDRO CARVALHO IZIDORO
-
Data: Apr 3, 2023
-
-
Show Abstract
-
This dissertation presents the Differential Transconductance Differences Amplifier (FDDTA) architecture based on CMOS inverters. Designed on a 130 nm CMOS process, it operates in weak inversion when supplied with 0.25 V. Furthermore, FDDTA does not require supplemental external calibration circuitry such as bias current or voltage sources as it relies on the distributed layout technique that intrinsically matches CMOS inverters. For analytical purposes, we performed a detailed investigation that describes all the concepts and the entire functioning of the FDDTA architecture.
|
|
6
|
-
NATALIA SÁNCHEZ SÁNCHEZ
-
Topological Location System using Computer Vision
-
Advisor : GIOVANI BERNARDES VITOR
-
COMMITTEE MEMBERS :
-
RUBEN DARIO HERNANDEZ BELENO
-
GIOVANI BERNARDES VITOR
-
RAFAEL FRANCISCO DOS SANTOS
-
Data: Apr 20, 2023
-
-
Show Abstract
-
This paper presents an innovative methodology that uses computer vision techniques to perform the topological localization of an autonomous vehicle. The great advantage of this technique is that it eliminates the use of GPS or any continuous time position sensor, which can significantly increase safety in regions where location sensors are limited or absent. The methodology consists of building a topological map of the region of interest, where the points of interest are defined. To do this, several images of each coordinate are collected and go through filters and processing to form a georeferenced image bank. From there, the system receives as input a video, where the images are compared with the images in the image bank, using the SURF algorithm, to define if there is a correspondence with the coordinates of interest. If a match is identified, the algorithm defines the location of the vehicle on the topological map. The results of the experiments performed show a 91.5\% accuracy rate in detecting the points of interest within the topological map, indicating that this methodology can complement the navigation system of an autonomous vehicle efficiently and accurately.
|
|
7
|
-
LUCAS GOMES DE ALMEIDA
-
Driver’s Behavior Classification in Vehicular Communication Networks for Commercial Vehicles
-
Advisor : BRUNO TARDIOLE KUEHNE
-
COMMITTEE MEMBERS :
-
STEPHAN REIFF-MARGANIEC
-
BRUNO TARDIOLE KUEHNE
-
EDVARD MARTINS DE OLIVEIRA
-
OTÁVIO DE SOUZA MARTINS GOMES
-
Data: May 26, 2023
-
-
Show Abstract
-
Vehicles are becoming more intelligent and connected due to the demand for faster, efficient, and safer transportation. For this transformation, it was necessary to increase the amount of data transferred between electronic modules in the vehicular network since it is vital for an intelligent system’s decision-making process. Hundreds of messages travel all the time in a vehicle, creating opportunities for analysis and development of new functions to assist the driver’s decision. Given this scenario, the dissertation presents the results of research to characterize driving styles of drivers using available information in vehicular communication network. This master thesis focuses on the process of information extraction from a vehicular network, analysis of the extracted features, and driver classification based on the extracted data. The study aims to identify aggressive driving behavior using real-world data collected from five different trucks running for a period of three months. The driver scoring method used in this study dynamically identifies aggressive driving behavior during predefined time windows by calculating jerk derived from the acquired data. In addition, the K-Means clustering technique was explored to group different behaviors into data clusters. Chapter 2 provides a comprehensive overview of the theoretical framework necessary for the successful development of this thesis. Chapter 3 details the process of data extraction from real and uncontrolled environments, including the steps taken to extract and refine the data. Chapter 4 focuses on the study of features extracted from the preprocessed data, and Chapter 5 presents two methods for identifying or grouping the data into clusters. The results obtained from this study have advanced the state-of-the-art of driver behavior classification and have proven to be satisfactory. The thesis addresses the gap in the literature by using data from real and uncontrolled environments, which required preprocessing before analysis. Furthermore, the study represents one of the pioneering studies conducted on commercial vehicles in an uncontrolled environment. In conclusion, this thesis provides insights into the development of driver behavior classification models using real-world data. Future research can build upon the techniques presented in this study and further refine the classification models. The thesis also addresses the threats to validity that were mitigated and provides recommendations for future research.
|
|
8
|
-
DANIEL PAIVA FERNANDES
-
Cost-efficient blockchain application to secure data transmission in heterogeneous FANETs
-
Advisor : JEREMIAS BARBOSA MACHADO
-
COMMITTEE MEMBERS :
-
SIDNEY NASCIMENTO GIVIGI JUNIOR
-
JEREMIAS BARBOSA MACHADO
-
RODRIGO MAXIMIANO ANTUNES DE ALMEIDA
-
SERGIO RONALDO BARROS DOS SANTOS
-
Data: Jun 26, 2023
-
-
Show Abstract
-
The development of vehicular networks has found a more fertile scenario with the advancement of ultra-reliable low latency communications (URLLC), deployment of fifth generation (5G) networks worldwide, empowerment of edge computing and adopting “Internet of Things” solutions in smart cities. To guarantee the success of these networks, it is essential to ensure that the communication process is reliable, safe from malicious actions, and that the solution has low computational complexity and energy consumption. Among vehicular networks that can take advantage of these new technologies are FANETs (Flying Ad-Hoc Networks), which can play a critical role in rescue missions and reconnaissance of risk areas. These networks need a solution that guarantees transparency, security and fault tolerance in a decentralised way to function correctly. Therefore, the present work proposes a proof-of-concept solution to ensure crash-fault tolerant communication in emulated heterogeneous Flying Ad-Hoc Networks (FANETs) using the Proof of Elapsed Time (PoET) consensus algorithm.
|
|
9
|
-
Bianca da Rocha Bartolomei
-
A proposal to support urban planning based on the use of data generated in the elaboration of public policy instruments
-
Advisor : MELISE MARIA VEIGA DE PAULA
-
COMMITTEE MEMBERS :
-
JOSE MARIA NAZAR DAVID
-
ADLER DINIZ DE SOUZA
-
MELISE MARIA VEIGA DE PAULA
-
VANESSA CRISTINA OLIVEIRA DE SOUZA
-
Data: Jun 30, 2023
-
-
Show Abstract
-
One of the existing goals in the 2030 Agenda for Sustainable Development is to increase inclusive and sustainable urbanization. Achieving this goal is already considered a challenge, since in Brazil many cities have already gone through, and still go through, a process of expansion and urbanization. In this context, the concept of urban planning emerges, understood as one of the ways to systematize this process, since it allows a better allocation of financial and human resources, in addition to defining actions and objectives in favor of solving collective problems. For this, urban policy instruments are defined, closely related to urban planning, since urban policy instruments are the tools and mechanisms used to implement planning and achieve the goals established for the city. The objective of the research presented in this master's thesis is to investigate solutions that support decision-making in the context of urban planning. For this, data collected in projects for the elaboration and updating of urban policy instruments were analyzed so that these data could be used in the elaboration of a solution. The methodology used was the Design Science Research Methodology (DSRM), the artifact proposed and developed was a decision support system in the form of a visualization panel of information about the spatial composition of a municipality in the interior of Minas Gerais . For this, concepts of geographic data analysis and information visualization were used. The panel was evaluated by a group of potential users and the hypothesis that the use of data generated by urban policy instruments can help municipal urban planning was corroborated by the responses obtained. With the study, it was possible to highlight the importance of the data considered, the potential of the proposed artifact, in addition to identifying opportunities for future work.
|
|
10
|
-
Iago Felicio Dornelas
-
Application of Laban's basic effort actions in an interactive 2D tool to support choreographic composition
-
Advisor : RODRIGO DUARTE SEABRA
-
COMMITTEE MEMBERS :
-
LINA MARIA GARCES RODRIGUEZ
-
LUCIANA APARECIDA MARTINEZ ZAINA
-
RODRIGO DUARTE SEABRA
-
Data: Jul 12, 2023
-
-
Show Abstract
-
The planning of movement in space by choreographers is crucial in choreographic composition, requiring a complex cognitive effort to transform an abstract product into a visual representation. Different means, from symbols and notations to digital tools, have been used to record and simulate movements. However, due to the specific nature of dance and its lack of availability as a technical training in Brazil, the methods consolidated over time, such as the concepts developed by choreographer Rudolf Laban, are not widespread or accessible to professional and amateur choreographers. We thus developed the Move Note tool, which allowed the participants in this research to explore dancers' trajectories through abstract animations. The tool made it possible to apply effects to the dancers' displacements, providing an innovative approach to represent Laban's basic effort actions in a two-dimensional environment. The development of the tool was based on an extensive bibliographic review, analysis of the state of the art and a survey on potential users. In order to investigate whether the application of Laban's concepts in an interactive tool could support choreographic composition, evaluations of users' experiences were carried out, adapted from the TAM (Technology Acceptance Model) and TTF (Task-Technology Fit) models. The results indicated that the tool developed was able to provide adequate support, since the satisfaction rates obtained in the analyses, together with the positive comments from the participants, evidenced the contribution provided by the tool. This work presents contributions both in terms of discussion about the interpretation of the data collected and reflection on the practical relevance of the research theme. Additionally, it introduces to the academic community a model of representation of Laban's basic effort actions in a two-dimensional environment, thus expanding the possibilities of research and application of these concepts to the fields of dance and technology.
|
|
11
|
-
MOISÉS PINHEIRO SOUZA
-
ProtCool 2.0: a client/server model for a docking protocol generator and molecular dynamics simulations in protein-ligand complex.
-
Advisor : CARLOS HENRIQUE DA SILVEIRA
-
COMMITTEE MEMBERS :
-
KARINA DOS SANTOS MACHADO
-
CARLOS HENRIQUE DA SILVEIRA
-
RODRIGO APARECIDO DA SILVA BRAGA
-
Data: Jul 18, 2023
-
-
Show Abstract
-
The COVID-19 pandemic has made it clear the high demand for computational systems that expedite the discovery of new drugs. In this regard, understanding the dynamic behavior of biomolecular complexes is crucial. Techniques involving molecular dynamics simulations of these complexes have increasingly been used to accelerate the identification of better drug candidates. However, the preparation of such simulations is highly complex, and their numerous details are not always adequately emphasized, compromising their reproducibility and reusability. To address this, the ProtCool tool was proposed—a protocol generator focused on integrating docking and molecular dynamics of protein-ligand complexes. In its initial version, this tool was restricted to the user's local environment. This work presents version 2.0 of ProtCool, developed under a clientserver model with a web interface. The aim is to fill the gaps left by the previous version, enhancing the software in three fundamental aspects: making it multi-platform, enabling access to multiple users, and making the tool more intuitive. The development of a userfriendly interface allows this new version to expand its scope of use to inexperienced or novice researchers in computational chemistry. ProtCool 2.0 does not execute the dynamics or perform result analyses; it is designed to be an expert in preparation based on workflows, following the programmed workflow and generating all the necessary configuration files for reliable execution of molecular dynamics on the user's computational setup. With the entire process being properly recorded, this allows for greater reproducibility and reusability of the preparations. Many of its functionalities are based on the adaptation of well-known tools from the literature in the field of molecular dynamics simulations. ProtCool 2.0 was developed using best practices and software engineering processes. Its client-server architecture implemented under the web standard enables it to be cross-platform and multi-user, providing benefits in availability and performance. It features a minimalist graphical interface with interactive resources that ensure user safety in correctly filling out their study parameters, thus preventing errors. To demonstrate its use, ProtCool 2.0 underwent a case study on the simulation of the interaction between acetylcholinesterase and galantamine, used in the treatment of Alzheimer's disease, which allowed for validation through replication of a simulation certified by peers in an international publication. The preparation of the simulation successfully enabled the execution of reliable molecular dynamics, reproducing the expected results. It is expected that this tool will not only bring greater speed, reproducibility, and reusability to molecular dynamics preparations but also contribute to smoothing the learning curve for these simulations in computational chemistry.
|
|
12
|
-
THIAGO MOREIRA DE FREITAS
-
Genetic Algorithms Applied to the Vehicle Routing Problem with Multiple Depots
-
Advisor : RAFAEL FRANCISCO DOS SANTOS
-
COMMITTEE MEMBERS :
-
FERNANDO BERNARDES DE OLIVEIRA
-
RAFAEL FRANCISCO DOS SANTOS
-
SANDRO CARVALHO IZIDORO
-
Data: Aug 30, 2023
-
-
Show Abstract
-
The Vehicle Routing Problem (VRP) has wide applications in logistics and transportation with great economic importance. VRP is a generalization of a large number of routing problems, which consist of finding the optimal number of routes, leaving a single depot, to serve a set of customers, minimizing routing costs and meeting a set of constraints. The Multi Depot Vehicle Routing Problem (MDVRP) is an extension of VRP, in which there is more than one depot distributed in a given geographic area. The rest of the problem is identical to VRP. There are several methods for solving MDVRP such as exact techniques, approximate algorithms and heuristics. Genetic Algorithms (GAs) are meta-heuristics widely used to find solutions to the MDVRP problem due to the stochastic characteristics of GAs and the efficiency in solving combinatorial problems and, for this reason, they were selected to be applied in this work. The developed algorithm was tested using instances present in the literature and compared with existing methodologies, in which the genetic algorithm found good results and the work contributed to the technique of selecting customers who can exchange between deposits. The results achieved show that this algorithm can be evaluated in real projects, making it possible to improve the operation of projects that face this type of problem, reducing transportation costs, distance, delivery time, services, among other benefits.
|
|
13
|
-
FERNANDO HIDEKI TAKENAKA
-
Review Summarizer Using TextRank and Topic Modeling
-
Advisor : LAERCIO AUGUSTO BALDOCHI JUNIOR
-
COMMITTEE MEMBERS :
-
ISABELA NEVES DRUMMOND
-
LAERCIO AUGUSTO BALDOCHI JUNIOR
-
RAFAEL DUARTE COELHO DOS SANTOS
-
Data: Aug 31, 2023
-
-
Show Abstract
-
Over the past decade, the Internet has changed the way people work, shop and socialize. Those changes resulted in the increase of User Generated Content (UGC) such as: ratings, reviews, wikis, and videos. UCG contains relevant information for decision-making, especially with regard to the acquisition of goods and services. However, the large volume and dispersion of this content makes it difficult to obtain relevant information. Text summarization appears as a way to make this content more accessible to people. A summary A can be considered better than another B when A is shorter than B while maintaining the same content relevance, or when A, despite being longer, presents more relevant content. Analyzing the literature, we observed that it is possible to produce better quality summaries than those produced by algorithms that correspond to the state of the art in text summarization. We present a multilingual automatic text summarizer that combines and extends the algorithms Latent Dirichlet Allocation (LDA) and TextRank. Our approach, when compared to the state of the art, generates better text summaries in terms of size and content relevance.
|
|
14
|
-
THIAGO SALES FREIRE LUZ
-
Analysis and comparison of ensemble classification algorithms on the exoplanet discovery
-
Advisor : ENIO ROBERTO RIBEIRO
-
COMMITTEE MEMBERS :
-
ENIO ROBERTO RIBEIRO
-
ROBERTO SILVA NETTO
-
RODRIGO APARECIDO DA SILVA BRAGA
-
SANDRO CARVALHO IZIDORO
-
Data: Sep 28, 2023
-
-
Show Abstract
-
Exoplanets are planets discovered outside our solar system. Their discovery happens because of scientific work with telescopes such as the Kepler. The data collected by Kepler is known as Kepler Object of Interest. Machine Learning algorithms are trained to classify these data into exoplanets or non-exoplanets. An Ensemble Algorithm is a type of Machine Learning technique that combines the prediction performance of two or more algorithms to gain an improved final prediction. The current works on exoplanet identification use mostly traditional non-Ensemble algorithms. Therefore, research that uses Ensemble algorithms for exoplanet identification is scarce. This paper performs a comparison among some Ensemble algorithms on the exoplanet identification process. Each algorithm is implemented with a set of different values for its parameters and executed multiple times. All executions are performed with the cross-validation method. A confusion matrix is created for each algorithm implementation. The results of each confusion matrix provided data to evaluate the following algorithm’s performance metrics: accuracy, sensitivity, specificity, precision, and F1 score. The Ensemble algorithms achieved an average performance of more than 80% in all metrics. Changing the default values of the Ensemble algorithms parameters improved their predictive performance. The algorithm with the best performance is Stacking. In summary, the Ensemble algorithms have great potential to improve exoplanet prediction. The Stacking algorithm achieved a higher performance than the other algorithms. This aspect is discussed in the text. The results of this work show that it is reasonable to increase the use of Ensemble algorithms. The reason is their high prediction performance to improve exoplanet identification.
|
|
15
|
-
RENATO FIGUEIREDO FRADE
-
Temporal and spatial characterization of street robberies contrasting pre-pandemic and pandemic contexts
-
Advisor : CARLOS HENRIQUE DA SILVEIRA
-
COMMITTEE MEMBERS :
-
ADRIANO VELASQUE WERHLI
-
ALEXANDRE CARLOS BRANDAO RAMOS
-
CARLOS HENRIQUE DA SILVEIRA
-
Data: Dec 9, 2023
-
-
Show Abstract
-
A multidisciplinary study conducted in Minas Gerais investigated the temporal dynamics of street robberies, analyzing both pre-pandemic and pandemic periods. Utilizing data from the Military Police, time series data were examined at various scales, including hourly, daily, 10-day intervals, and monthly, employing advanced statistical methods such as spectral frequency analysis, autocorrelations, and decomposition techniques.
|
|
16
|
-
BRENO DE OLIVEIRA RENÓ
-
Conf-eHealth - An Architecture for Developing eHealth Applications with Trustworthiness
-
Advisor : EDVARD MARTINS DE OLIVEIRA
-
COMMITTEE MEMBERS :
-
HENRIQUE YOSHIKAZU SHISHIDO
-
BRUNO TARDIOLE KUEHNE
-
EDVARD MARTINS DE OLIVEIRA
-
Data: Dec 13, 2023
-
-
Show Abstract
-
This work presents the Conf-eHealth architecture, which aims to be a reference model for developing eHealth applications with trustworthiness. The architecture is relevant due to the needs of technologies related to patient monitoring. Initially, a systematic literature review was carried out on the state of the art of eHealth applications and the main challenges faced in their development. After the literature review, the proposed reference architecture is described. The conception of needs, quality attributes and the methodology used to build the architecture are described and subsequently the architecture is presented through the concept of architectural visions.
In order to guarantee the desired quality attributes, the work presents the evaluation of the proposed architecture in two parts. First an architectural evaluation was carried out based on the Software Architecture Analysis Method, resulting in a conceptual understanding of the architecture's ability to encompass quality attributes. Afterwards, an experiment was carried out, involving the construction of a mobile application according to the component mapped in the proposed architecture, explaining the possibility of implementing this component using a real database. The results show that the application is capable of handling the data received and assisting in decision making accurately. Finally, the conclusions of the work are presented, highlighting the results achieved, the importance of the Conf-eHealth architecture for the advancement of the area and indicating future work.
|
|
17
|
-
ANDRE LUIZ ALVES DIAS
-
Detection of Dissatisfaction in Public Servants with Artificial Intelligence
-
Advisor : CARLOS HENRIQUE VALERIO DE MORAES
-
COMMITTEE MEMBERS :
-
AHMED ALI ABDALLA ESMIN
-
CARLOS HENRIQUE VALERIO DE MORAES
-
JOAO PAULO REUS RODRIGUES LEITE
-
Data: Dec 15, 2023
-
-
Show Abstract
-
This work highlights a comprehensive investigation into the application of Artificial Intelligence (A.I.) in human resources management, with a specific focus on identifying employee dissatisfaction through machine learning approaches. The research included a review of scientific articles discussing both the implementation of A.I. in the context of human resources and the use of machine learning techniques to detect cases of turnover/attrition, along with the relationship between dissatisfaction and turnover/attrition cases. To assess these approaches, four validated public databases were selected. Three of them contained fictional employee data, and one contained real employee turnover data. Each database underwent a process of textual field factorization, followed by analyses to highlight the data distributions in each set. In conducting the research, different machine learning approaches were applied to each of the databases, aiming to verify the feasibility of identifying dissatisfaction through A.I. The techniques used included anomaly or novelty detection, classifiers, and optimized sets of classifiers. The results were quantified, revealing promising scores, with performances exceeding 90%. These results emphasize the overall effectiveness of machine learning in identifying employee dissatisfaction, demonstrating its potential for practical applications in the human resources environment.
|
|
18
|
-
CHARLY BRAGA VENTURA
-
Identification of miner and rust in coffee plants using digital image processing and convolutional neural networks
-
Advisor : SANDRO CARVALHO IZIDORO
-
COMMITTEE MEMBERS :
-
RODRIGO APARECIDO DA SILVA BRAGA
-
SANDRO CARVALHO IZIDORO
-
VALDETE MARIA GONÇALVES DE ALMEIDA
-
Data: Dec 15, 2023
-
-
Show Abstract
-
The world’s demand for coffee increases every year, reaching 178.5 million 60 kg bags in the period 2022-2023, an increase of 1.7% compared to the previous period 2021-2022. The total production of Brazil’s coffee harvest in 2022 was calculated at 50.92 million bags of 60 kg of processed coffee 2, thus making it the world’s largest producer of the product. With this production volume, there is a growing need to improve product quality due to the demands of national and international markets. However, pests such as leaf miner and rust cause extensive damage to coffee plantations, resulting in crop losses annually. Various methods and techniques have been developed and applied to assess the level of infestation and control of these pests. Among these techniques are the use of computer vision and convolutional neural networks (CNNs). Thus, the objective of this work was to develop computational tools to correctly identify the presence of pests, reducing evaluation time, evaluator error, and labor costs. The accuracies of these methods developed were between 99.67% and 97.00%. In addition, a tool was developed to quantify the degree of infestation, achieving an accuracy of 86.67%.
|
|
19
|
-
RAFFAEL CLEISSON DE CARVALHO
-
An analysis of the application of Nudge in public consultations
-
Advisor : MELISE MARIA VEIGA DE PAULA
-
COMMITTEE MEMBERS :
-
PABLO VIEIRA FLORENTINO
-
ADLER DINIZ DE SOUZA
-
MELISE MARIA VEIGA DE PAULA
-
Data: Dec 18, 2023
-
-
Show Abstract
-
Public consultation is a direct democracy device that aims to allow the population to participate in decision-making in various areas of public policy. In Brazil, the Fiscal Responsibility Law establishes that a public consultation must precede the preparation of the public budget, which aims to regulate the state's financial revenue and expenditure activities for one year. This example of public consultation is called Participatory Budgeting. However, population engagement is still a problem for several reasons. Therefore, it is important to investigate strategies that increase citizen participation. In some scenarios, Nudges theory can be used. Nudges are small, low-investment modifications made to a person's environment to change their behavior. The application of Nudge in this research aims to encourage citizens to participate in a public consultation context. To enable analysis, it was part of the scope of this research to propose and develop a tool. The application was analyzed in a real scenario which allowed us to conclude that the results with the application of Nudge were adequate.
|
|