https://repositorio.ufu.br/handle/123456789/5142 2026-08-28T22:09:06Z 2026-08-28T22:09:06Z https://repositorio.ufu.br/handle/123456789/49977 2026-08-28T18:33:39Z 2026-08-11T00:00:00Z

Title: Avaliação de sistema de recomendação para atividades pedagógicas no moodle

2026-08-11T00:00:00Z https://repositorio.ufu.br/handle/123456789/49975 2026-08-28T18:24:20Z 2026-08-07T00:00:00Z

Title: Análise comparativa de modelos de aprendizado de máquina para detecção de ataques DDoS em ambientes IoT com seleção de features baseada em SHAP

2026-08-07T00:00:00Z https://repositorio.ufu.br/handle/123456789/49941 2026-08-28T12:54:19Z 2026-07-28T00:00:00Z

Title: Estudo de caso do uso de um sistema de gerenciamento de banco de dados para detecção de cyberbullying Abstract: The ease of access to social media has intensified episodes of cyberbullying among users, a practice that negatively affects victims’ lives. Given this problem, this work aims to evaluate the applicability of PGVector (an extension for creating VDBMSs) in processes for detecting messages potentially associated with cyberbullying, combining semantic similarity search with a supervised machine learning model. To this end, data from the X platform (formerly Twitter) were collected and submitted to a method structured in six steps: textual preprocessing, vector representation, sampling, storage in PostgreSQL with the PGVector extension, classification, and evaluation of results. In the classification step, three embedding generation models were compared — SBERT, BERT, and FastText —, each combined with the kNN algorithm executed directly in the VDBMS. The results showed that SBERT achieved the best performance (k = 9, f1-score of 0.77), outperforming BERT (k = 7, f1-score of 0.65) and FastText (k = 5, f1-score of 0.59), confirming that embeddings optimized for similarity comparison are more suitable for this type of task. It is concluded that PostgreSQL, through PGVector, is capable of acting as a complete similarity-based classification environment, validating its applicability for cyberbullying detection.

2026-07-28T00:00:00Z https://repositorio.ufu.br/handle/123456789/49903 2026-08-28T06:19:59Z 2026-07-22T00:00:00Z

Title: Portal Web para otimização de processos e expansão digital: um caso aplicado à Ice Point Sorveteria Abstract: Small and Medium Enterprises (SMEs) face significant challenges in managing their logistics operations, often relying on manual processes prone to human error and rework. This work presents the development and implementation of a comprehensive web system for Ice Point Sorveteria, structured under the MVC (Model-View-Controller) architectural pattern. The application encompasses an e-commerce platform for the end consumer (B2C) and an internal management dashboard, aiming to mitigate errors in renting limited stock and automate logistics calculations. The technological infrastructure was built using the Vue.js framework for the frontend, NestJS for the backend, and Supabase acting as a Backend-as-a-Service (BaaS) for PostgreSQL persistence and secure authentication (SSO). Among the validated automated business rules are the wholesale minimum threshold (over eighty units), the twelve-hour logistical interval for fleet sanitization, and dynamic freight calculation integrated with the Google Maps API. Software quality is ensured by an automated unit testing suite using the Jest framework, covering eleven critical business rule scenarios, complemented by a Continuous Integration and Continuous Delivery (CI/CD) pipeline via GitHub Actions that automates verification and deployment with every code update. The results indicate the complete elimination of deficit dispatches and a traceable standardization of revenues and scheduling, consolidating the platform as a technological milestone in the company’s delivery operations.

2026-07-22T00:00:00Z