Dissertações defendidas no âmbito do Programa. https://tedebc.ufma.br/jspui/handle/tede/1314 2026-04-13T10:10:12Z 2026-04-13T10:10:12Z PASSOS, Arthur Marinho dos https://tedebc.ufma.br/jspui/handle/tede/6813 2026-03-03T13:06:58Z 2025-09-16T00:00:00Z

Título: Desenvolvimento de artefatos de software para apoiar a avaliação de bibliotecas digitais móveis focando em experiência do usuário Autor: PASSOS, Arthur Marinho dos Primeiro orientador: RIVERO CABREJOS, Luis Jorge Enrique Abstract: This dissertation was developed within the scope of a project by the Open University of the Unified Health System (UNASUS/UFMA), whose goal is to expand access to educational resources for healthcare professionals through a mobile digital library. The migration of content previously restricted to a web environment to mobile devices revealed important challenges related to usability, accessibility, and user experience, especially considering the demands of use in real professional practice contexts. To address these challenges, the research adopted a structured approach composed of exploratory studies, which enabled an understanding of the usage context and the identification of limitations present in existing mobile digital libraries, and a Systematic Literature Review (SLR), which consolidated the main quality attributes relevant to this type of application. These results supported the development of BiblioCheck, an inspection checklist designed to evaluate usability, accessibility, and quality of use in mobile digital libraries. The checklist was assessed by inspectors and experts, who analyzed its clarity, applicability, and ability to identify real issues. The results demonstrated that BiblioCheck is effective in supporting systematic evaluations, facilitating the identification of critical problems and guiding improvements in the design and functionality of these systems. Qualitative feedback from the evaluators contributed to refining the instrument, increasing its precision and alignment with the specificities of mobile use. In the end, the dissertation delivers three main contributions: an updated mapping of quality attributes for mobile digital libraries, the creation of a structured and validated instrument for inspecting these systems, and practical recommendations that can guide future design, evaluation, and development initiatives. The work therefore contributes to overcoming existing gaps in the field and offers direct support for enhancing the user experience in mobile digital libraries, including the UNASUS/UFMA project and similar initiatives. Instituição: Universidade Federal do Maranhão Tipo do documento: Dissertação

2025-09-16T00:00:00Z NUNES, Kennedy Anderson Mendes https://tedebc.ufma.br/jspui/handle/tede/6791 2026-02-24T12:54:59Z 2025-09-18T00:00:00Z

Título: Desenvolvimento de artefatos para a avaliação de dashboards de sistema de apoio a tomada de decisão focando em usabilidade e experiência do usuário Autor: NUNES, Kennedy Anderson Mendes Primeiro orientador: CABREJOS, Luis Jorge Enrique Rivero Abstract: Well-designed dashboards synthesize complex data, allowing users to quickly identify trends and patterns. To achieve their goals, these dashboards must be easy to use, enhancing users’ ability to understand, interact with, and extract insights from the presented data. This paper highlights the importance of dashboards in supporting decision-making, emphasizing the crucial role of user experience (UX) and usability in the effectiveness of such systems. In recent years, dashboards have become increasingly popular, with a significant rise in the number of these tools emerging in the market. This growth has encouraged developers to create their own tools and has also attracted the attention of researchers to this area. Despite their prominence, there is a need to propose technologies that can support developers and development teams in the process of designing and evaluating quality dashboards. While dashboards are on the rise, the technologies aimed at ensuring their quality and user satisfaction have not kept pace. The main objective of this master’s research is to propose quality attributes related to usability and user experience that can be incorporated during the dashboard development process. Following a literature review on dashboard quality attributes, a checklist was developed to assess the usability aspects of these systems. The checklist facilitates a structured and simple identification of usability issues, even by inexperienced users, serving as a robust evaluation tool based on quality attributes validated by previous literature. Additionally, an aggregated set of Design Patterns was proposed and associated with the checklist verification items. Both the inspection checklist and the design patterns were applied to evaluate and redesign dashboards proposed within an information system for decision-making at the multinational energy company Equatorial Energia. The results of this experience suggest the feasibility of considering these quality attributes to improve the ease of use of dashboards. To evaluate the UDASHBOARD checklist and the DP-DASHBOARD design patterns, an experimental study was conducted. The results indicated that participants considered the technology useful for identifying defects in dashboards; however, its ease of use could be improved. Furthermore, it was found that participants’ experience did not significantly influence the effectiveness and efficiency outcomes of the technique. Regarding the design patterns, results showed that, overall, the technology is easy to understand and useful in supporting dashboard design. However, the visual examples of some patterns could be improved to make them clearer for developers. Instituição: Universidade Federal do Maranhão Tipo do documento: Dissertação

2025-09-18T00:00:00Z SILVA, Karla Felícia Carvalho da https://tedebc.ufma.br/jspui/handle/tede/6610 2025-11-13T11:50:12Z 2025-08-29T00:00:00Z

Título: Impacto das técnicas de equilíbrio de classes na privacidade: uma análise dos ataques de inferência de associação Autor: SILVA, Karla Felícia Carvalho da Primeiro orientador: COUTINHO, Luciano Reis Abstract: The use of machine learning models trained on unbalanced datasets with sensitive information has raised significant privacy concerns. One major issue is vulnerability to Membership Inference Attacks (MIAs), which aim to discover whether a particular data point was part of a model’s training set. In this context, it has been largely overlooked whether class imbalance handling methods — a prerequisite for training standard supervised classifiers on such data — exacerbate the disclosure of sensitive information. To address this gap, we conducted a series of empirical MIA evaluations on real-world imbalanced datasets to investigate this question. Our results show that techniques employed to balance the dataset for classification can significantly improve the success rate of MIA attacks. This has clear implications for researchers developing protective measures against such attacks, especially in scenarios with imbalanced datasets. Instituição: Universidade Federal do Maranhão Tipo do documento: Dissertação

2025-08-29T00:00:00Z SOUZA, João Pedro Gomes de https://tedebc.ufma.br/jspui/handle/tede/6516 2025-09-29T17:26:21Z 2025-09-22T00:00:00Z

Título: Detecção de fissuras em micro tomografia de concreto reforçado por fibras utilizando redes neurais profundas Autor: SOUZA, João Pedro Gomes de Primeiro orientador: SILVA, Aristófanes Corrêa Abstract: Fiber-reinforced concrete is a crucial material for civil construction, and monitoring its healh is an important task for preserving structures and preventing accidents and financial losses. Among the various non-destructive concrete health monitoring methods, Computed Microtomography (Micro-CT) image analysis stands out as an inexpensive method free from noise and external interference. However, manual inspection of cracks in Micro-CT images is subjective and requires significant human effort. Deep learning algorithms have been used extensively for concrete cracks automatic detection, but only in surface regions. Detecting cracks all fiber-reinforced concrete structure, including inside in high-resolution Micro-CT volumes, remains a challenge. Therefore, this work proposes a framework for automatic crack detection explicitly for fiber-reinforced concrete Micro-CT images, combining super-resolution-based preprocessing, Detection Transformers (DETR), and committee-based post-processing to improve crack detection reliability and reduce false positives. The model was trained on a new publicly available dataset, the FIRECON dataset, which consists of 4,064 images annotated by an expert, achieving metrics of 86.098% Intersection over Union (IoU), 89.37% Precision, 83.26% Recall, 84.99% F1-Score, and 44.69% Average Precision (AP). This framework, therefore, significantly reduces analysis, human effort, and improves consistency compared to manual methods used in previous studies. The results demonstrate the potential of deep learning to aid image analysis in concrete health assessments, understanding of the cracking mechanisms of fiber-reinforced concrete, and provides a valuable tool for the development of durable, high-performance engineering materials. Instituição: Universidade Federal do Maranhão Tipo do documento: Dissertação

2025-09-22T00:00:00Z