Isabel Rosseti

Predicting soccer match outcomes is valuable for coaches and players engaged in strategic planning, for enhancing fan engagement, and for the economic, media, and journalistic activities that surround the sport. The task requires analyzing large volumes of data, including team statistics, player performance metrics, and historical results. Although player and team interactions intuitively play a pivotal role in determining outcomes, automated prediction methods based on statistical models and machine learning often overlook this dimension. This paper proposes modeling such interactions using graph representations and investigates heterogeneous graph neural networks for extracting latent interaction features. We further propose integrating these graph-based features with tabular features derived solely from past match outcomes to assess the contribution of interaction information to predictive performance. Specifically, we employ state-of-the-art heterogeneous graph neural network models to extract latent interaction features and fuse them with the observed tabular features to train an XGBoost classifier. We evaluate the approach on eight datasets drawn from the top leagues of England, Spain, France, Brazil, Germany, Italy, the Netherlands, and Argentina. Across the analyzed seasons, the classifier using heterogeneous graph-generated features outperformed the tabular-only classifier in a single year, while the classifier combining both feature sets achieved competitive performance throughout. Notably, however, the combined model did not consistently improve upon the tabular-only baseline in terms of Ranked Probability Score (RPS): in most seasons it matched rather than exceeded the baseline. These findings indicate that graph-based interaction features carry a complementary signal that can be incorporated without degrading predictive performance, while motivating further investigation into how this signal can be more effectively exploited.

Criminal activity is a major challenge for governments in large urban centers worldwide. Within Smart City initiatives, ensuring urban safety is a key priority. However, developing and applying data-driven approaches to reduce urban crime remains an open, yet important, issue. This paper addresses the Routing of Police Vehicles in Large Urban Centers (RVP-Urb) problem, which aims to maximize crime coverage while considering resource constraints. The proposed solution, called PROTECTOR, is a heuristic based on the Greedy Randomized Adaptive Search Procedure (GRASP), designed to generate preventive patrol routes that enhance police visibility in crime hotspots. Unlike existing methods, PROTECTOR mirrors the structure of Brazilian police departments by dividing urban areas into smaller zones and generating both intra- and inter-zone routes. The heuristic uses real-world crime hotspot data over defined time intervals to create daptive patrol plans that can be updated as hotspot patterns change. Experimental results using the RolRoute-DS ataset demonstrate that PROTECTOR performs effectively, producing near-optimal routes comparable to those enerated by exact mathematical models while maintaining computational efficiency.

The Bounded-Diameter Minimum Spanning Tree Problem (BDMSTP) is defined as follows: Given a graph G where each edge xy has a positive cost w _xy the goal is to find an optimal spanning tree of G whose diameter does not exceed a prescribed positive integer D ≤2. Applications of the BDMSTP appear in telecommunication network and fiber optic projects, data compression problems, and distributed systems design. This work revisits search methods for the BDMSTP and proposes alternative combinations of local search moves, intending to select efficient sets of moves and improve the quality of the solutions found in the literature. For small 50-and 100-node instances from the widely-used OR-Library, we obtain exact solutions whose optimal values were so far unknown. Having solved these easier cases, we can then concentrate our efforts on solving more challenging OR-Library graph instances with 250, 500, and 1000 nodes. The process of selecting efficient sets of search moves is done by encapsulating different local search versions (based on the Variable Neighborhood Descent method) in an ILS approach. Some new neighborhoods are also proposed. Computational tests show that these neighborhoods allowed the proposed ILS approach to obtain better results than those presented in the literature.

This paper presents an approach to ease dataflow analysis in Computational Science and Engineering (CSE) applications that invoke serverless functions. By capturing provenance data while executing CSE applications within a serverless environment, the approach organizes this information into an integrated database that users can query at runtime. Since serverless platforms typically lack native support for provenance tracking, the proposed solution helps users understand and analyze the behavior and outcomes of their CSE applications. We detail the main features of the approach as implemented in the DENETHOR tool and evaluate its effectiveness with a real-world CSE application. The results demonstrate that DENETHOR enhances analytical capabilities via its provenance database.

The concept of Smart Cities has gained relevance, especially in the last decade, due to the availability of data associated with cities, e.g., car traffic, public transportation, crime data, etc. The purpose of using these data is to improve the services offered to the citizens. Most of these applications manipulate spatiotemporal data. These data are processed in a dataflow that starts with the collection, integration, and aggregation and ends with visualization. This way, specialized data services for smart city applications are most welcome. However, many of the existing data services in this context, are either specific to a particular application/domain or do not consider the entire data life cycle. In this article, we present Hurricane, a dataflow-oriented data service for smart city applications. Hurricane executes multiple dataflows to gather, pre-process, integrate, and public data. Hurricane was evaluated with an application in the area of public security and results reinforced the importance of this type of data service.

Recently, hybrid metaheuristics have become a trend in operations research. A successful example combines the Greedy Randomized Adaptive Search Procedures (GRASP) and data mining techniques, where frequent patterns found in high-quality solutions can lead to an efficient exploration of the search space, along with a significant reduction of computational time. In this work, a GRASP-based state-of-the-art heuristic for the Minimum Latency Problem (MLP) is improved by means of data mining techniques for two MLP variants. Computational experiments showed that the approaches with data mining were able to match or improve the solution quality for a large number of instances, together with a substantial reduction of running time. In addition, 88 new cost values of solutions are introduced into the literature. To support our results, tests of statistical significance, impact of using mined patterns, equal time comparisons and time-to-target plots are provided.