Mestrado em Informática

URI Permanente para esta coleção

http://repositorio.ufes.br/handle/10/17627

Nível: Mestrado Acadêmico
Ano de início:
Conceito atual na CAPES:
Ato normativo:
Periodicidade de seleção:
Área(s) de concentração:
Url do curso:

Navegar

Agora exibindo 1 - 20 de 235

A general purpose conceptual model for crowdsourcing projects
(Universidade Federal do Espírito Santo, 2018-08-07) Assis Neto, Fábio Ribeiro de; Santos, Celso Alberto Saibel; Pereira Filho, José Gonçalves; Santos, Vaninha Vieira dos
Crowdsourcing is a method that employs people to process input data to solve a computationally complex problem, such as generating a large dataset of annotated images, audio transcriptions or video scene descriptions. In this approach, people select tasks and produce individual results according to a list of steps that leads to an efficient solution. Then, every single result must be collected, interpreted, and integrated by a platform or system supporting the crowdsourcing process. Considering that the first studies regarding this field of study were done by means of experimenting and observing platforms and systems that popularized this method. Given that, pertinent questions are: (i) besides platforms and systems, what are the essential elements that compose a crowdsourcing project?; (ii) how the humans are guided through the steps for solving the proposed problem?; (iii) how to ensure the quality of the results produced in all stages of a project?; and (iv) are there guidelines or models in the state-of-the-art to help newcomers creating their own projects? Aiming to answer these questions, a systematic literature review based on the viewpoint of individuals that seek to solve problems using the crowdsourcing method was performed during this MSc dissertation, thus stating that this domain lacks general guidelines and models that leverage the essential elements of this kind of project, as well as its relationships and descriptions. Therefore, this dissertation aims to present a set of tendencies and guidelines related to crowdsourcing project and propose a general-purpose conceptual model that represent the essential elements involved and how they relate in this kind of project, in a way that is possible to instantiate and expand this model to supply specific needs of individual projects. Furthermore, the proposed model was applied during the planning and execution stages of a crowdsourcing project designed locally.
A model-driven approach to the conceptual modeling of situations: from specification to validation
(Universidade Federal do Espírito Santo, 2015-10-28) Sobral, Vinicius Marchandt; Costa, Patrícia Dockhorn; Almeida, João Paulo Andrade; Pires, Luís Ferreira; Pereira Filho, José Gonçalves
The modeling of situation types for context-aware applications, also called situationaware applications, is, on the one hand, a key task to the proper functioning of those applications. On the other hand, it is also a hard task given the complexity and the wide range of possible situation types. Aiming at facilitating the representation of those types of situations at design-time, the Situation Modeling Language (SML) was created. This language is based partially on rich ontological theories of conceptual modeling and is accompanied by a platform for situation-detection at runtime. Despite the benefits of the availability of this suitable infrastructure, the definition of situation types, being a non-trivial task, can still pose problems that are hardly detected by modelers by manual model inspection. This thesis aims at improving and facilitating the definition of situation types in SML by proposing: (i) the integration between the language and the ontological theories of conceptual modeling by using the OntoUML language, with the purpose of increasing the expressivity of situation type models; and (ii) an approach for the validation of situation type models using a lightweight formal method, aiming at increasing the correspondence between the created models’ instances and the modeler’s intentions. Both the integration and the validation are implemented in a tool for specification, verification and validation of ontologically-enriched situation types.
A model-predictive motion planner for the IARA autonomous car
(Universidade Federal do Espírito Santo, 2017-11-23) Cardoso, Vinícius Brito; Badue, Claudine; Souza, Alberto Ferreira de; Santos, Thiago Oliveira dos; Osório, Fernando Santos
In this work, we present the Model-Predictive Motion Planner (MPMP) of the Intelligent Autonomous Robotic Automobile (IARA). IARA is a fully autonomous car that uses a path planner to compute a path from its current position to the desired destination. Using this path, the current position, a goal in the path and a map, IARAs MPMP is able to compute smooth trajectories from its current position to the goal in less than 50 ms. MPMP computes the poses of these trajectories so that they follow the path closely and, at the same time, are at a safe distance from occasional obstacles. Our experiments have shown that MPMP is able to compute trajectories that follow precisely a path produced by a human driver (distance of 0.15m in average) while smoothly driving IARA at speeds of up to 32.4 km/h (9 m/s).
A multiagent architecture to multilingual interaction
(Universidade Federal do Espírito Santo, 2017-08-11) França, Eduardo dos Santos; Tavares, Orivaldo de Lira; Menezes, Crediné Silva de; Castro Junior, Alberto Nogueira
This dissertation presents a virtual environment specially constructed with resources to aid in the communication between speakers of different languages, aimed at enabling interaction in textual environments between pairs or groups of people, within learning communities that do not have a common language. In this environment there is a digital agent (or bot) that uses communication strategies to assist the interlocutors in the understanding of messages in foreign languages, and may also define technical terms, regionalisms and expressions according to dictionaries and services used. The environment resources were focused on synchronous interactions, but can also be used in asynchronous interactions with some adaptations. In addition to supporting such communication between speakers of different languages, interactions are stored for research on the effects of agents, strategies, and resources on communication
A neural-based model predictive control to tackle steering delay of the IARA autonomous car
(Universidade Federal do Espírito Santo, 2017-09-04) Guidolini, Rânik; Gonçalves, Claudine Santos Badue; Santos, Thiago Oliveira dos; Wolf, Denis Fernando
In this work, we propose a Neural Based Model Predictive Control (N-MPC) approach to tackle delays in the steering plant of autonomous cars. We examined the N-MPC approach as an alternative for the implementation of the Intelligent and Autonomous Robotic Automobile (IARA) steering control subsystem. For that, we compared the standard solution, based on the Proportional Integral Derivative (PID) control approach, with the N-MPC approach. The PID steering control subsystem works well in IARA for speeds of up to 25 km/h. However, above this speed, IARA’s Steering Plant delays are too high to allow proper operation with a PID approach. We tried and modeled the IARA’s Steering Plant using a neural network and employed this neural model in the N-MPC approach. The N-MPC approach outperformed the PID approach by reducing the impact of IARA’s Steering Plant delays and allowing the autonomous operation of IARA at speeds of up to 37 km/h – an increase of 48% in the maximum stable speed.
A novel cooperative algorithm for clustering large databases with sampling
(Universidade Federal do Espírito Santo, 2012-07-30) Fabris, Fábio; Varejão, Flávio Miguel; Alvarenga, Arlindo Gomes de; Barbosa, Hélio José Corrêa; Rodrigues, Alexandre Loureiro
Clustering is a recurrent task in data mining. The application of traditional heuristics techniques in large sets of data is not easy. They tend to have at least quadratic complexity with respect to the number of points, yielding prohibitive run times or low quality solutions. The most common approach to tackle this problem is to use weaker, more randomized algorithms with lower complexities to solve the clustering problem. This work proposes a novel approach for performing this task, allowing traditional, stronger algorithms to work on a sample of the data, chosen in such a way that the overall clustering is considered good.
A reactive GRASP algorithm for the multi-depot vehicle routing problem with time windows
(Universidade Federal do Espírito Santo, 2022-07-19) Souza, Israel Pereira de; Boeres, Maria Claudia Silva; https://orcid.org/0000-0001-9801-2410; http://lattes.cnpq.br/0528154281423964; Ochi, Luiz Satoru; Amaral, Andre Renato Sales; http://lattes.cnpq.br/4695002674556067
The vehicle routing problem (VRP) is a well know hard to solve problem in literature. In this work, we describe a reactive greedy randomized adaptive search procedures algorithm, for short, reactive GRASP, using a variable neighborhood descent (VND) algorithm as local search procedure to solve the multi-depot vehicle routing problem (MDVRP) and multi depot vehicle routing problem with time windows (MDVRPTW). This algorithm, called RGRASP+VND, combines four distinct local search procedures and a clustering technique. The Cordeau et al. dataset, a widely well known MDVRP benchmark, is considered for the experimental tests. RGRASP+VND achieves better results on most small instances and a lower average solution cost for all instances on the experimental tests when compared to the earlier GRASP approaches in the MDVRP literature. RGRASP+VND results are also compared with the state-of-the-art of MDVRP and MDVRPTW.
AAP - Um Ambiente para Aprendizagem de Programação
(Universidade Federal do Espírito Santo, 2014-12-04) Chagas, Lucinéia Barbosa da Costa; Menezes, Crediné Silva de; Tavares, Orivaldo de Lira; Nevado, Rosane Aragón de; Oliveira, Marcia Gonçalves de
Several research groups around the world have studied the process of learning how to program a computer. The learning of programming continues to be a complex problem because of the multiple skills and multiple stages required to the programming process. In the view of many students, introductory programming courses are considered difficult, this results in poor performance and many failures. In this research, we are interested in the creation of a supportive environment for learning programming in its several stages. Thus we define a virtual environment with computing resources to support the implementation of pedagogical architectures useful for learning programming. We present the elements of these architectures (objectives, strategies and digital resources) and their validations based on the apprentices’ experiences.
Abordagem do problema de programação de grade horária sujeito a restrições utilizando coloração de grafos
(Universidade Federal do Espírito Santo, 2007-11-12) Bello, Geraldo Simonetti; Rangel, Maria Cristina; Boeres, Maria Cláudia Silva; Oliveira, Elias Silva de; Ochi, Luiz Satoru
This work addresses the Class/Teacher Timetabling Problem and Graph Coloring Problem, presents features, describes the main approaches proposed in the literature for the solution of these two problems, including the reformulation of the basic version of the Class/Teacher Timetabling Problem as a Graph Coloring Problem, and uses the algorithms on this approach for their resolution. A Class/Teacher Timetabling Problem with additional constraints to those found in the basic version is chosen in the literature and a model that extends the correspondence between the two problems is developed, also contemplating additional constraints. Adaptations in a Tabu Search algorithm for Graph Coloring described in the literature and in the existing implementation in C language are proposed, in order to allow the resolution of the problem with the additional constraints considered. The chosen problem is reformulated as a Graph Coloring Problem and the modified algorithm is used for its resolution. The computational results for a set of instances related to the problem are presented and the effect of this reformulation in the results is verified, comparing them with those in the literature, which does not use this reformulation.
ADALiN – ambiente digital de aprendizagem de línguas
(Universidade Federal do Espírito Santo, 2019-02-19) Araujo, Raissa Arruda de; Tavares, Orivaldo de Lira; https://orcid.org/0000-0003-4734-3193; http://lattes.cnpq.br/5998785809434157; https://orcid.org/0000-0002-9085-7475; http://lattes.cnpq.br/4564768528153271; Menezes, Credine Silva de; https://orcid.org/0000-0002-2709-7135; http://lattes.cnpq.br/5454484334693909; Castro Junior, Alberto Nogueira de ; https://orcid.org/0000-0002-1752-8667; http://lattes.cnpq.br/5919189481858271
This dissertation presents the design, development, implementation and evaluation of a "digital environment" for learning natural languages, entitled ADALin. The novelty is that this environment is designed to serve a community of learners, without separating borders, in the sense that they reach different levels of proficiency, using five (5) abilities: reading, writing, speaking, listening, translating. This digital environment allows the use of Digital Learning Structures, created by teachers to be used for individual or collaborative learning with apprentices of different proficiency levels. Each learner's learning path is freely defined by himself. When the learner needs help choosing the next task to be developed, the environment recommends the activities, structured in the respective EDA, that are appropriate to the current level and proficiency of that learner.
Agregação de classificadores neurais via integral de Choquet com respeito a uma medida fuzzy
(Universidade Federal do Espírito Santo, 2016-07-15) Pacheco, André Georghton Cardoso; Kröhling, Renato Antônio; Boeres, Maria Claudia Silva; Papa, João Paulo
Data classification appears in many real-world problems, e.g., recognition of image patterns, differentiation among species of plants, classifying between benign and malignant tumors, among others. Many of these problems present data patterns, which are difficult to be identified, thus requiring more advanced techniques to be solved. Over the last few years, various classification algorithms have been developed to address these problems, but there is no classifier able to be the best choice in all situations. So, the concept of ensemble systems arise, which more than one methodology is used together to solve a particular problem. As a simple and effective methodology, ensemble of classifiers have been applied in several classification problems, aiming to improve performance and increase reliability of the final result. However, in order to improve the classification accuracy, an affective aggregation of classifiers must be performed. In this work, we present two contributions: first, we describe three classifiers based on neural networks, a multilayer feedforward trained by Levenberg-Marquardt algorithm; an extreme learning machine (ELM); and a discriminative restricted Boltmann machine (DRBM). Furthermore, we use conventional classifier k-nearest neighbors (KNN). Next, we propose an aggregation methodology to ensemble of classifiers using Choquet integral with respect to a fuzzy measure obtained by principal component analysis (PCA). Then, we apply this methodology to aggregate the classifiers performed to conventional benchmarks, for large database and the results are promising.
Algoritmo A-Estrela de estado híbrido aplicado à navegação autônoma de veículos
(Universidade Federal do Espírito Santo, 2013-08-28) Gonçalves, Michael André; Santos, Thiago Oliveira dos; Souza, Alberto Ferreira de; Gonçalves, Claudine Santos Badue; Aguiar, Edilson de; Chaimowicz, Luiz
In this work, we investigated the use of A-star algorithms (A*) with hybrid state in au-tonomous navigation of vehicles in a three-dimensional space. We have modeled the vehicle position (origin), the goal and other points of interest in the world (states) as nodes of a graph. The cost of navigating between these nodes were modeled as edges of the graph, and a variant of the A* algorithm was used to choose the best path be-tween origin and goal. In order to be able to avoid obstacles and achieve fast algorithm, we used a combination of two heuristics to estimate the cost of the current node to the goal node: one considering only the obstacles and without the limitation of rotation of the vehicle, and its dual disregarding the obstacles and with limited cinematic R³. We implemented the proposed navigation solution and incorporated it to the framework of robotics CARMEN as a navigation module for autonomous vehicles. Our module interacts with other existing modules (interface modules with sensors, mapping, local-ization, etc.) by means of message exchanging. It enables practical use of the algo-rithm. Results of experiments performed on IARA (Intelligent Robotic Autonomous Automobile - autonomous drive car developed in UFES) showed the viability of using the algorithm in simple and structured environments, such as roads, as well as in un-structured and complex environments, such as parking lots and unpaved areas,
Algoritmo genético híbrido aplicado ao problema de agrupamento de dados
(Universidade Federal do Espírito Santo, 2009-08-31) Alckmin, Danuza Prado de Faria; Varejão, Flávio Miguel; Martins, Simone de Lima; Boeres, Maria Claudia Silva
Clustering is a task that divides a data set in subgroups aiming that elements associated to one exactly group are more similar between themselves than elements of other groups. Organizing data in groups make it possible to identify similarities and differences between them, to extract useful information and conclusions regarding the data features. Clustering may be considered an optimization problem because it is intended to find the best combination of partitions among all possible combinations. An approach that can be applied to solve the clustering problem is the use of metaheuristics, which are procedures capable of escaping from local optima, once the use of exact methods is computationally infeasible. However, the majority of the metaheurísticas applied to clustering problem is not scalable for real or commercial bases. They are more effective for smaller instances of the problem trated. The computational cost necessary to calculate the solutions becomes greater in larger instances of the problem. For this reason, hybrid procedures that explore the combination of metaheuristics represent a promising approach for solving the clustering problem. This work shows a proposal of a Hybrid Genetic Clustering Algorithm that associates the process of global search to a local search heuristic and also initializes the population by different grouping techniques. Such improvements aim to direct the search for solutions next to the global optimal one. An experimental evaluation with real and synthetic databases is performed aiming to verify if the proposed approach presents an improvement in relation to the other evaluated algorithms. The result of this analysis shows that the proposed algorithm presents a better performance in four among the six evaluated algorithms. In addition, an analysis of the execution time shows that the execution time of our proposal is feasible, even though it is considerably longer than the execution times of the fast convergence algorithms.
Ambiente inteligente e colaborativo para apoio à produção acadêmica: esclarecimento de dúvidas
(Universidade Federal do Espírito Santo, 2015-02-28) Vale, Karla Samantha Bezerra; Menezes, Crediné Silva de; Tavares, Orivaldo de Lira; Gava, Tânia Barbosa Salles; Castro Junior, Alberto Nogueira de
The development of academic work, as we know, requires great effort of the authors, largely operational nature, occupying hours of work that could be dedicated to the analysis and creation activities. Among these activities we can mention the collection of information on the subject to be searched, the management items read or to be read, or the search for publications on the theme. Yet we know that many of these efforts can be mitigated through initiatives of cooperation and use of computational tools. There are reputable tools that support some of these steps, however the academic needs to combine two or more of these tools to attend his needs. Looking help to alleviate this problem we designed, at LIEd, an Intelligent and Collaborative Environment, which combines artificial intelligence techniques in order to provide a collaborative environment able to support computationally, in an integrated way, some of the essential steps of the academic production. This paper presents a subsystem of the environment, Document Retrieval, which provides information directly and automatically from questions in natural language, considering the context of a project and the documentary evidence indicated by the user during their interactions.
Ambiente inteligente e colaborativo para apoio à produção acadêmica: sistema de recomendação
(Universidade Federal do Espírito Santo, 2015-09-04) Cruz, Patrícia de Oliveira; Menezes, Crediné Silva de; Tavares, Orivaldo de Lira; Castro Júnior, Alberto Nogueira de
abstract
Ambientes virtuais de aprendizagem: a inclusão autônoma dos aprendizes deficientes visuais
(Universidade Federal do Espírito Santo, 2015-06-05) Lima, Jefferson Ribeiro de; Tavares, Orivaldo de Lira; Nevado, Rosane Aragón de; Menezes, Crediné Silva de
This article presents a proposal for development of inclusive digital resources in virtual learning environments in order to support the conduct of the visually impaired learner in performing the tasks and interactions that occur in the platform only by spoken instructions via microphone attached to the computer. Making fully autonomous visually impaired learner for navigation and communication in the learning environment without the need for assistance of third parties or external
An architecture for end-to-end network slicing in multiple data centers using tableless source routing
(Universidade Federal do Espírito Santo, 2020-07-30) Valentim, Rodolfo Vieira; Villaca, Rodolfo da Silva; https://orcid.org/0000000280513978; http://lattes.cnpq.br/3755692723547807; https://orcid.org/0000-0002-7702-2991; http://lattes.cnpq.br/1345691729490868; Pasquini, Rafael; https://orcid.org/0000-0002-8781-3914; http://lattes.cnpq.br/6428800770934048; Martinello, Magnos; https://orcid.org/0000-0002-8111-1719; http://lattes.cnpq.br/7471111924336519
Telecommunication Service Providers (TSP) need to offer a wide range of services to their customers. The network functions virtualization (NFV), software-defined networks (SDN), and network slicing paradigms facilitate the deployment of these services by managing virtualized resources in a software-driven way. These paradigms have gained traction with the growth of new services that have different business demands, because their composition enhance programmability and innovation on top of COTS equipment and open-source tools. However, there are cases when these services are composed by functions hosted in geographically distant data centers (or clouds) in different domains. Telecommunication networks are almost ubiquitous, but multiple TSPs provide access to these networks, and each TSP has a footprint focused on a specific region. This market fragmentation makes it challenging to deploy cost-effective network services spanning multiple clouds and few works in the literature focus on the mechanisms for implementing the stitching of network slices from multiples service providers or infrastructures. In this context, this work presents a solution for implementing an end-to-end network slicing solution using tableless source routing. We propose the use of tableless source routing to specify a set of physical links, sub-networks, and network functions using a single piece of information without stacking protocol headers. At the same time, it improves scalability in the network core by reducing the number of forwarding states. Our proposal of the architecture has two main concerns. First, it needs to enable tableless source routing in data centers in a flexible, easy, and programmatic way, keeping compatibility with legacy applications hosted in these data centers. The second task and our major contribution is to perform the stitching of slicing segments in each data center in a way that abstracts from operators the huge amount of work required to perform these tasks. As a proof-of-concept, we implemented prototypes of the proposed solution with cutting edge cloud technologies, such as OpenFlow, OpenStack, and Open vSwitch. The results of functional and performance tests showed that the proposed solution enables end-to-end network slicing in a low-cost, efficient and flexible manner. Moreover, our proposal provides mechanisms to the NFV orchestrator that allow fine-grained traffic engineering decisions to optimize the selection of network paths and connections between domains.
An experimental study of the geometric and algebraiz multigrid strategies
(Universidade Federal do Espírito Santo, 2016-09-30) Carrion, Marcelo Torres Pereira; Catabriga, Lucia; Boeres, Maria Claudia Silva; Zambom, Eduardo; Santos, Isaac Pinheiro dos; Reis Junior, Neyval Costa
abstract
An ontological theory of the electrocardiogram with applications
(Universidade Federal do Espírito Santo, 2009-05-13) Gonçalves, Bernardo; Guizzardi, Giancarlo; Pereira Filho, José Gonçalves; Fonseca, Frederico; Almeida, João Paulo Andrade
The fields of Medical- and Bio-informatics are bearing witness of the application of the discipline of Formal Ontology to the representation of biomedical entities and (re-)organization of medical terminologies also in view of advancing electronic health records (EHR). In this context, the electrocardiogram (ECG) defines one of the prominent kinds of biomedical data. As a vital sign, it is an important piece in the composition of the EHR of today, as likely in the EHR of the future. This thesis introduces an ontological analysis of the ECG grounded in the Unified Foundational Ontology (UFO) and axiomatized in First-Order Logic (FOL). With the goal of investigating the phenomena underlying this cardiological exam, we deal with the sub-domains of human heart electrophysiology and anatomy. We then outline an ECG ontology meant to represent what the ECG is on both sides of the patient and of the physician. The ontology is implemented in the semantic web technology OWL with its SWRL extension. The ECG Ontology makes use of basic relations standardized in the OBO Relation Ontology for the biomedical domain. In addition, it takes inspiration in the Foundational Model of Anatomy (FMA) and applies the Ontology of Functions (OF). Besides the ECG ontological theory itself, two applications of the ECG Ontology are also presented here. The first one is concerned with the off-line integration of ECG data standards, a relevant endeavor for the progress of Medical Informatics. The second one in turn comprises a reasoning-based web system that can be used to offer support for interactive learning in electrocardiography / heart electrophysiology. Overall, we also reflect on the ECG Ontology as well as on its two applications to provide evidence for benefits achieved with the employment of methodological principles - in terms of both ontological foundations and ontology engineering - in building a domain ontology.
An Ontology to support Knowledge Management Solutions for Human-Computer Interaction Design
(Universidade Federal do Espírito Santo, 2021-05-19) Castro, Murillo Vasconcelos Henriques Bittencourt; Barcellos, Monalessa Perini; https://orcid.org/0000-0002-6225-9478; http://lattes.cnpq.br/8826584877205264; https://orcid.org/0000-0002-9976-0696; http://lattes.cnpq.br/5598991250064438; Conte, Tayana Uchôa; https://orcid.org/0000-0001-6436-3773; http://lattes.cnpq.br/6682919653508224; Souza, Vítor Estevão Silva; https://orcid.org/0000000318695704; http://lattes.cnpq.br/2762374760685577
Developing interactive systems is a challenging task. It involves concerns related to the human-computer interaction (HCI), such as usability and user experience. Therefore, HCI design must be addressed when developing such systems. HCI design often involves people with different backgrounds, technical languages, terms and knowledge, what makes communication and knowledge transfer a challenging issue. In this scenario, knowledge management can support understanding concepts from different knowledge areas and help learn from previous experiences. Knowledge management has supported HCI design mainly to improve product quality and reduce effort and time spent on design activities. However, there is a need for simpler and more practical knowledge management solutions to support HCI design. In addition, the lack of a common conceptualization about HCI design has been one of the main challenges to be addressed. This leads to semantic interoperability problems, such as ambiguity and imprecision when interpreting shared information, and hampers communication and knowledge transfer. Aiming to provide a well-founded conceptualization about HCI design domain in the context of the development of interactive systems, this work proposes HCIDO (Human-Computer Interaction Design Ontology). HCIDO is a reference ontology of the Human-Computer Interaction Ontology Network (HCI-ON) and is also connected to the Software Engineering Ontology Network (SEON), allowing for the reuse of concepts related to Software Engineering and HCI aspects, such as requirements, code, interactive systems and users, as well as making them connected to design aspects. HCIDO was evaluated through verification and validation techniques. Moreover, a computational tool was developed using HCIDO as a reference model, illustrating how the ontology can be applied to support knowledge management solutions in HCI design. The tool supports knowledge management activities (e.g., knowledge capture, representation, storage, retrieval, use and evaluation) in the HCI design of interactive systems by allowing HCI designers to annotate structured information about design choices in design artifacts shared with HCI design stakeholders.

Navegar

Navegando Mestrado em Informática por Título

Resultados por página

Opções de Ordenação