https://repositorio.uema.br/jspui/handle/123456789/2202 Tue, 09 Jun 2026 05:12:14 GMT 2026-06-09T05:12:14Z http://repositorio.uema.br:80/retrieve/8bb1dea7-85ac-4502-8c56-9ea6cc16c94a/Eng. de Computação.jpg https://repositorio.uema.br/jspui/handle/123456789/2202 https://repositorio.uema.br/jspui/handle/123456789/6179 Título: São João do Maranhão: análise de tópicos e sentimentos em tweets e notícias Abstact: The growing percentage of the population with internet access in Brazil means that social media such as X is being used more frequently and, consequently, text data generated by its users is becoming more available. Therefore, it is interesting to collect text samples from users and analyze opinions on a given topic. In this study, sentiment analysis and topic modeling were performed using data from X and G1 about São João do Maranhão during the 2023 and 2024 festivities. The objective was to identify the impression left by the event on the population of Maranhão. To this end, web scraping techniques were applied to collect data. In total, 1,756 tweets and 125 news items were collected. After collection, the texts underwent pre-processing methods to allow sentiment classiĄcation and topic modeling. To identify the topics covered in social media, BERTopic was used, which was successful in identifying the main topics covered by Internet users, such as: São João da Thay, Bumba Meu Boi, art and folklore. Regarding sentiment analysis, it was found that the vast majority of tweets and articles published by G1 were positive, conĄrming the broad satisfaction of users in relation to São João do Maranhão. Fri, 14 Feb 2025 00:00:00 GMT https://repositorio.uema.br/jspui/handle/123456789/6179 2025-02-14T00:00:00Z https://repositorio.uema.br/jspui/handle/123456789/5414 Título: Marandu Hub: desenvolvimento de um sistema low-code para gerenciamento de espaços de coworking com flutterflow Abstact: This study presents the development of a reservation management system for Marandu Hub, a coworking space maintained by Marandu – UEMA Agency for Innovation and Entrepreneurship. With the expansion of collaborative environments and the need for more practical solutions to organize the use of these spaces, particularly in academic contexts, the proposal to automate the room scheduling process emerged. The platform was built with FlutterFlow, a low-code tool that allows for the visual, rapid, and scalable creation of applications. The system replaces outdated manual reservation methods, facilitating room usage and preventing scheduling conflicts. Implemented features include secure login, administrator control panels, automatic email notifications, and dynamic report generation. The research adopted an applied approach with descriptive and qualitative methods, seeking solutions that effectively met the institution's needs. Key technologies used were Firebase (for user authentication and database), Node.js (to create an API for Excel reports), and EmailJS (for sending automated emails). The system was designed to serve distinct user profiles and was adjusted to function well on both mobile phones and computers. To evaluate the user experience, the System Usability Scale (SUS) was applied, resulting in an average score of 94.5 points, which demonstrates that the application was well-received. The system is currently in use and has contributed to more efficient management, serving as an example of how low-code tools can be valuable for modernizing institutional processes with limited resources. Keywords: reservation system; coworking; FlutterFlow; Thu, 24 Jul 2025 00:00:00 GMT https://repositorio.uema.br/jspui/handle/123456789/5414 2025-07-24T00:00:00Z https://repositorio.uema.br/jspui/handle/123456789/5145 Título: ECOLIFE: Utilização de IA Generativa no desenvolvimento de um jogo sério sobre sustentabilidade Abstact: The growing advancement of artificial intelligence (AI) technologies has caused major changes in society, enabling the most diverse tasks to be carried out with the help of these tools, especially with generative AI. Therefore, there are many areas that can benefit from its use, such as game development. This work investigates the use of two generative artificial intelligences, ChatGPT and Gemini, during the creation of a serious game in a quiz format aimed at environmental sustainability, using commands on these platforms to generate questions and answers for the application, in three different themes: 5R’s, sustainable practices and A3P Agenda. The results obtained show that the use of these AIs as an aid to this type of problem is useful, with well- evaluated scenarios being generated that are consistent with what was proposed, despite the presence of some errors. Sat, 09 Nov 2024 00:00:00 GMT https://repositorio.uema.br/jspui/handle/123456789/5145 2024-11-09T00:00:00Z https://repositorio.uema.br/jspui/handle/123456789/5099 Título: Método chan-vese de segmentação aplicado na localização de lesões de câncer melanoma e não melanoma Abstact: Skin cancer is the most common of all cancers and the increased incidence is due in part to the people´s behavior in relation to sun expose. In Brazil, non-melanoma skin cancer is the most frequent in most regions. Dermatoscopy is the main kind of exam for the diagnosis of dermatological skin diseases. The computer aided medical diagnosis – CAD, has become increasingly common, such as the diagnosis of skin lesions, where the techniques for automatic extraction of their contours becomes crucial. Usually, the system starts by pre-processing the image, in other words removing undesired artifacts such as hair, freckles or shading effects. Next, the system performs a segmentation step to identify the lesion boundaries. The idea of CAD can be applied to all kinds of obtaining image, including conventional radiography, computed tomography, magnetic resonance, ultrasound and nuclear medicine. The computer's response can be useful, since the diagnosis of the health professional is based on subjective assessment, subject to intraand inter-personal variation and low image quality, eye fatigue or distraction. This study aims to present an analysis of the Chan-Vese model for segmentation of dermatological images of melanoma and non-melanoma types. The Chan-Vese segmentation model is based on the region growing technique and assets boundary model. The model showed a good result of segmented images in relation to the number of images used. The procedure of segmentation analysis was performed according to the results that had already been collected from database in which the images were acquired Wed, 06 Jul 2016 00:00:00 GMT https://repositorio.uema.br/jspui/handle/123456789/5099 2016-07-06T00:00:00Z