The growth of basketball viewership and the expansion of sports betting in Brazil have created new opportunities for data-driven sports analytics. However, research on predictive modeling for Brazilian basketball remains scarce, largely due to the absence of publicly available structured datasets. This study addresses that gap by introducing the first comprehensive dataset dedicated to NBB (acronym for Novo Basquete Brasil in Portuguese), compiled through the systematic collection and organization of match-level statistics. In addition, we evaluate the performance of well-established machine learning algorithms for game outcome prediction, employing a novel feature engineering strategy based on team statistics from the last K games and a sliding window training approach. The XGBoost consistently delivered the best results, especially when trained with larger datasets. In sliding window experiments, it achieved excellent performance (F1-Score = 0.89) with 128 training samples, reinforcing the importance of data volume for reliable prediction. These findings offer a valuable baseline for future research and demonstrate the potential of statistical modeling in enhancing understanding and strategic decision-making in Brazilian basketball.

Over recent years, several sophisticated Process-Aware Information Systems (PAIS) have been proposed for managing business processes and automating large-scale scientific (e-Science) processes. Much of this success is due to their ability to provide generic functionality for modeling, execution and monitoring processes. These functionalities work well when process execution follows a well-behaved path towards achieving the models objectives. However, exceptions and anomalous situations that fall outside of the well-behaved execution path still pose a significant challenge to PAIS. The treatment for such exceptions usually involves interventions in systems by human operators, which result in significant additional cost for businesses. In this paper, we introduce a cost-aware recovery composition method that is able to find and follow recovery paths that reduce the cost of exception handling. From a practical point of view, our proposal reduces complexity and the need for manual interventions to handle exceptions. Finally, the feasibility of recovery mechanism is discussed from its implementation into WED-flow framework.

Microservices is a modern architectural style in which developers decomposes a software system into many services loose coupled with small responsibilities. Given its inherent complexity, many frameworks have been proposed in order to support developers in microservices. However, due to its particularities, the whole process of choosing the most appropriate framework for developers' needs is a time-consuming and challenging task. In this paper, we present a qualitative study that compares both KumuluzEE and Spring Cloud & NetFlix OSS frameworks through functional and non-functional requirements. We tested each framework by developing a hypothetical scenario with each of them. Our results show that although the KumuluzEE supports few characteristics of the microservices architecture, it is easier to use, especially, for newcomers. Instead, the Spring Cloud & NetFlix OSS is suitable for large-scale systems and experienced development teams, and it holds a higher number of the architecture characteristics. However, learnability for newcomers is low even though the framework provides a substantial documentation.