RESUMEN
The objective of the URBANITE project is to design an open-data, open-source, smart-city framework to enhance the decision-making processes in European cities. The framework's basis is a robust and user-friendly simulation tool that is supplemented with several innovative service modules. One of the modules, a multi-output, machine-learning unit, is deployed on the simulation results, enabling city officials to more effectively analyse vast quantities of data, discern patterns and trends, and so facilitate advanced policy decisions. The city's decision makers define potential city scenarios, key performance indicators, and a utility function, while the module assists in identifying the policy that is best aligned with the stipulated constraints and preferences. One of the main improvements is a speeding up of the policy testing for the decision makers, reducing the time needed for one policy verification from 3 hours to around 10 seconds. The system was evaluated for Bilbao's Moyua area, where it suggested strategies that could result in a decrease in emissions of more than 5% CO2, NOx, PM in the selected area and a broader part of the city with a machine-learning accuracy of 91%. The system was therefore able to provide valuable insights into effective policies for restricting private traffic in specific districts and identifying the most advantageous times for these restrictions.
RESUMEN
The COVID-19 pandemic affected the whole world, but not all countries were impacted equally. This opens the question of what factors can explain the initial faster spread in some countries compared to others. Many such factors are overshadowed by the effect of the countermeasures, so we studied the early phases of the infection when countermeasures had not yet taken place. We collected the most diverse dataset of potentially relevant factors and infection metrics to date for this task. Using it, we show the importance of different factors and factor categories as determined by both statistical methods and machine learning (ML) feature selection (FS) approaches. Factors related to culture (e.g., individualism, openness), development, and travel proved the most important. A more thorough factor analysis was then made using a novel rule discovery algorithm. We also show how interconnected these factors are and caution against relying on ML analysis in isolation. Importantly, we explore potential pitfalls found in the methodology of similar work and demonstrate their impact on COVID-19 data analysis. Our best models using the decision tree classifier can predict the infection class with roughly 80% accuracy.