https://repositorio.uema.br/jspui/handle/123456789/1885 2026-07-01T00:43:17Z 2026-07-01T00:43:17Z https://repositorio.uema.br/jspui/handle/123456789/6213 2026-06-25T18:10:54Z 2025-01-01T00:00:00Z

Título: Universidade VOIP: caso de usuário posto de combustível Abstact: Telephony continues to be a fundamental element in the operations of any company today, and this is for very clear and practical reasons. It not only ensures direct and reliable communication between employees, clients, and partnersŮhelping to solve problems immediately, close deals without wasting time, and provide instant customer supportŮbut it also perfectly Ąts the well-known business maxim: ŞTime is money.Ť For many executives and managers, investing in eicient telephony is synonymous with agility, as it allows teams to handle urgent demands quickly, avoiding delays that can be costly in terms of productivity and lost opportunities. Although the use of traditional telephony has declined due to the growth of messaging and internet-based calling applications, voice communication has not lost its relevance. What has occurred is a transformation in the medium used: technology has migrated from Ąxed lines and analog exchanges to digital platforms, which are more Ćexible and better adapted to the corporate environment. In recent years, the widespread adoption of high-speed internet has enabled voice to be transmitted as data packets, paving the way for Voice over IP (VoIP) solutions. This change marked a milestone for companies of all sizes, as it allowed cost reductions, integration of advanced features such as call recording and call queues, and the scalability necessary to support organizational growth. Despite the strong presence of communication applications in peopleŠs daily lives, in business contexts they still present limitations, such as the lack of integration with internal systems and the diiculty of maintaining security and contact privacy. For this reason, corporate telephony based on VoIP has consolidated itself as the most eicient alternative, combining cost savings, reliability, and intelligent call management features. In summary, telephony has not ceased to be important; on the contrary, it has evolved and adapted to new demands. Today, more than just a voice channel, it integrates with companiesŠ data networks and digital tools, becoming a fundamental part of modern technological infrastructure

2025-01-01T00:00:00Z https://repositorio.uema.br/jspui/handle/123456789/6187 2026-06-18T17:18:10Z 2026-01-01T00:00:00Z

Título: Desenvolvimento de aplicação web para rastreamento de aeronaves Abstact: This work presents the development of a web application designed for the monitoring and tracking of aircraft operated by the Tactical Air Center of the Military Police of Maranhão. The solution was built using ReactJS and integrates with a REST API to display real-time information on aircraft position, trajectory and operational data during missions. In addition to tracking, the application provides access to flight history and enables the management of missions, aircraft and users through an intuitive and responsive interface. The tests conducted demonstrated that the tool delivers stable performance, continuous data updates and good adherence to the operational needs of the CTA, highlighting the potential of web development as support for aerial operations.

2026-01-01T00:00:00Z https://repositorio.uema.br/jspui/handle/123456789/6186 2026-06-24T17:21:55Z 2026-02-05T00:00:00Z

Título: Automação de entrega de software com Docker e CI/CD: um estudo de caso experimental Abstact: Software development faces the critical challenge of ensuring delivery reliability amidst manual processes prone to failure. This work investigates the technical implementation of infrastructure automation as a mechanism to mitigate operational inconsistencies. Limited to a controlled academic experimental case study, a Continuous Integration and Continuous Delivery (CI/CD) pipeline was developed for a Python web application (FastAPI). The architecture utilized Docker containers to ensure environment parity and GitLab CI to orchestrate testing, build, security, and deploy stages on an EC2 instance in the AWS cloud. The analysis of results prioritized the stability metric over pure speed. The data showed that, while there was a 20.35% optimization in average Lead Time, the most significant gain was the elimination of process variability (drastic reduction in standard deviation), transforming delivery into a predictable routine. Additionally, the Shift-Left approach was technically validated using Docker Scout for vulnerability scanning. It is concluded that automation, even in smaller-scale scenarios, is essential to ensure the technical reproducibility, security, and stability of software engineering.

2026-02-05T00:00:00Z https://repositorio.uema.br/jspui/handle/123456789/6185 2026-06-12T14:10:22Z 2025-01-01T00:00:00Z

Título: Reduzindo alucinações e aprimorando a qualidade de respostas em sistemas alimentados por large language models através de re-ranking de documentos para aplicações em domínios específicos Abstact: This work addresses the problem of hallucinations in Large-Scale Language Models (LLMs) by improving Retrieval Augmented Generation (RAG) systems. Specifically, it proposes a document re-ranking model, based on BERT architecture, to refine the results retrieved in a RAG system, prioritizing more relevant documents and, consequently, mitigating the generation of incorrect or unfounded information (hallucinations) by LLMs. To train and evaluate the model, an innovative dataset was created from ten Undergraduate Thesis Projects (TCCs) from the State University of Maranhão (UEMA), using Natural Language Processing (NLP) techniques for text extraction and automatic generation of questions with different levels of relevance (scores 1, 3 and 5). The BERTimbau re-ranking model was trained to classify question and document pairs according to their relevance. Experimental results demonstrate that the model achieves 92% accuracy in relevance classification and an MRR of 0.7367, indicating a significant improvement in the ordering of retrieved documents compared to an approach without re-ranking (MRR of 0.4140). The qualitative analysis illustrates the model's ability to discern between different levels of semantic relationship between questions and documents. This work contributes to the advancement of the state of the art in RAG, providing an effective method to reduce hallucinations in LLM and improve the reliability of information generated in applications in the Portuguese language domain, specifically in the context of UEMA's final course projects

2025-01-01T00:00:00Z