Ronaldo dos Santos Mello

Data-intensive applications often combine different database models, such as relational, document and graph stores. This strategy, known as polyglot persistence, must decide which model is more suitable for each domain data. This is a complex decision in early design stages as it usually depends on factors such as the conceptual data structure and the expected workload. This ongoing PhD research proposes a novel methodology to support polyglot database design from conceptual schemas, workload estimates, and non-functional requirements. The proposal extracts structural and access-related features, applies preliminary decision rules, and records the decision rationale in a polyglot metadata catalog. Preliminary results with relational and document-oriented data models indicate that data model suitability depends on the interaction between schema structure, query pattern and nesting strategy.

The brModelo initiative began in 2005 with the purpose of providing semiautomated solutions to assist Database modeling education. Supported by the UFSC Database Group, the initiative comprises a set of tools developed over these 21 years, of which brModeloWeb is the most recent. Available since 2016, brModeloWeb reaches 10 years of existence with more than half a million registered users from universities and professionals that adopt it for both educational and real-world database design. This paper celebrates the successful adoption of brModeloWeb and presents a new extension for NoSQL logical design, introduced in the latest release, along with performance improvements. A usability evaluation conducted during a workshop at the Brazilian Database School (ERBD 2026) indicated that the tool was well received.

Like online games and the financial market, several applications require specific data management features such as large data volume support, data streaming, and the processing of thousands of OLTP transactions per second. In general, traditional relational databases are not suitable for these requirements. NewSQL is a new generation of databases that combines high scalability and availability with ACID support, being a promising solution for these kinds of applications. Although data partitioning is an essential feature for tuning relational databases, it is still an open issue for NewSQL databases. This paper proposes an automated approach for hybrid data partitioning that minimizes the number of distributed transactions and keeps the system well-balanced. In order to demonstrate its efficacy, we compare our solution with an optimal partitioning solution generated by a solver and a state-of-art baseline. The experiments show that the quality of the partitioning scheme is similar to the optional solution and overcomes the state-of-art approach in number of distributed transactions.

Due to the environmental impacts of greenhouse gas emissions from traditional combustion vehicles, governments worldwide are encouraging the transition to electric vehicles (EVs). However, as EV use increases, user-related charging challenges have become evident. To identify possible solutions to improve EV charging management from a user-centered perspective, a state-of-the-art study was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology. Optimization systems and artificial intelligence (AI) methods applied to decision-making were compared and a growing trend towards the implementation of artificial intelligence in current applications was identified. This study investigates in more depth the application of AI in multi-agent systems for energy management in EV charging. It provides a critical review of charging stations, focusing on aggregator-based models that operate within a multi-agent system in smart grids. This analysis adopts the vehicle owner’s perspective and considers the charging duration of the EV as a parameter. This article identifies significant gaps in how existing research addresses individual electric vehicle users, noting a lack of consideration for energy management and system connectivity to support EV recharging locations. This work presents solutions to these gaps by using aggregators and multi-agent systems to represent charging stations, facilitate user access, and improve energy management.

A Auditoria e a governança de dados em projetos com múltiplas fontes exigem rastreabilidade confiável, garantindo a qualidade e confiabilidade dos dados ao longo de seu ciclo de vida. A linhagem de dados surge como uma ferramenta essencial para mapear a origem, transformações e destinos dos dados. Este trabalho propõe uma abordagem para definir e implementar o percurso completo dos dados ao longo do tempo. Utilizou-se metadados padronizados pelo Dublin Core, enriquecidos para capturar e persistir o histórico de transformações. A abordagem inclui a modelagem de metadados específicos para rastrear operações de ETL (Extração, Transformação e Carga) e demonstrou ser viável em um projeto com dados reais.

Thousands of vessels operating worldwide may present issues related to anomalous trajectories, often characterized by unexpected changes in course, speed, or navigation through restricted areas. Manually detecting these anomalies is impractical, evidencing the need for automated support for maritime surveillance agents. Although Automatic Identification Systems (AIS) enhance situational awareness, they are insufficient for detecting trajectory anomalies. Existing approaches already use AIS data to detect anomalous vessel trajectories, but they do not consider contextual variables (e.g., oil spills) or the domain expertise of maritime surveillance agents. This paper introduces INSTRUCTOR, a solution that combines AIS data with clustering algorithms to group vessel trajectories based on similar directions and movement patterns. This clustering output assists experts in collaboratively identifying anomalous trajectories, a process validated through the collective judgment of multiple domain specialists. Experiments conducted in partnership with the Brazilian Navy corroborate the effectiveness of INSTRUCTOR in detecting anomalous maritime trajectories.

Decidir a melhor ferramenta para gestão de metadados é um desafio para equipes de projetos de dados. Este artigo apresenta uma análise de duas ferramentas de código aberto para esta finalidade, OpenMetadata e Amundsen, com enfoque no projeto Céos e seu escopo em dados governamentais. A metodologia OSSpal foi utilizada para avaliação e comparação a fim de garantir decisões assertivas e aumento na confiança da escolha de uma ferramenta. Os prós e contras de cada ferramenta no escopo do projeto Céos foram obtidos, com destaque para a OpenMetadata com score de 4.55 contra 3.75 do Amundsen.