Abstract
We present S+t-SNE, an adaptation of the t-SNE algorithm designed to handle infinite data streams. The core idea behind S+t-SNE is to update the t-SNE embedding incrementally as new data arrives, ensuring scalability and adaptability to handle streaming scenarios. By selecting the most important points at each step, the algorithm ensures scalability while keeping informative visualisations. By employing a blind method for drift management, the algorithm adjusts the embedding space, which facilitates the visualisation of evolving data dynamics. Our experimental evaluations demonstrate the effectiveness and efficiency of S+t-SNE, whilst highlighting its ability to capture patterns in a streaming scenario. We hope our approach offers researchers and practitioners a real-time tool for understanding and interpreting high-dimensional data.

Abstract
The complex trajectories of higher education students are deviations from the regular path due to delays in completing a degree, dropping out, taking breaks, or changing programmes. In this study, we investigated degree changing as a cause of complex student trajectories. We characterised cohorts of students who graduated with a complex trajectory and identified the characteristics that influenced the time to graduation. To support this predictive task, we employed machine learning techniques such as neural networks, support vector machines, and random forests. In addition, we used interpretable techniques such as decision trees to derive managerial insights that could prove useful to decision-makers. We validated the proposed methodology taking the University of Porto (Portugal) as case study. The results show that the time to degree (TTD) of students with and without complex trajectories was different. Moreover, the proposed models effectively predicted TTD, outperforming two benchmark models. The random forest model proved to be the best predictor. Finally, this study shows that the factors that best predict TTD are the median TTD and the admission regime of the programme of destination of transfer students, followed by the admission average of the previous programme. By identifying students who take longer to complete their studies, targeted interventions such as counselling and tutoring can be promoted, potentially improving completion rates and educational outcomes without having to use as many resources.

Abstract
Stopword lists, an essential resource for natural language processing and information retrieval, are often unavailable for low-resource languages. Creating these lists is time-consuming and expensive, making automated stopword detection a viable alternative. This paper introduces a novel stopword detection approach that exploits the topological properties of co-occurrence networks to identify function words. By leveraging the connectivity patterns of function words in these networks, the proposed approach aims to achieve higher precision compared to traditional frequency-based methods. To assess the effectiveness of the network-based approach, we constructed co-occurrence networks for Tetun and Emakhuwa (low-resourced languages), as well as English and Portuguese. We then compared the performance of this approach with traditional frequency-based methods. The results indicate that the network-based approach consistently outperforms traditional methods, with in-degree emerging as the most reliable indicator of function words. This finding suggests promising prospects for automatically generating stopword lists in other low-resource languages, paving the way for developing natural language processing tools for these linguistic contexts. © 2024 PROPOR. All Rights Reserved.

Abstract

Abstract
With the increasing adoption of electric motorcycles in urban environments, efficient energy management becomes essential to maximize the autonomy and sustainability of these vehicles. This study proposes the development of forecasting models to predict energy consumption and generation as means to optimize the charging of electric motorcycle batteries. Three models are explored in this work, namely multiple linear regression, LSTM (Long Short-Term Memory) neural networks, and XGBoost (Extreme Gradient Boosting). The performance of each model is assessed through various metrics. The results indicate that the LSTM model exhibited the best performance, particularly in identifying complex temporal patterns in solar radiation data. However, XGBoost also proved to be reasonable, while multiple linear regression was less satisfactory. The study discusses its limitations, such as the lack of deep refinement of model parameters, and future perspectives, including the exploration of other models and the implementation of strategies for predictive battery charging management.

Abstract
The automotive industry is witnessing a surge in the production of electric vehicles (EVs) driven by stringent emission regulations. Despite this growth, heavy-duty truck fleets, particularly in waste collection, remain predominantly combustion-based ones. Waste collection is critical in urban environments, presenting unique challenges due to confined operational regions. One alternative to increase EVs in waste collection is to substitute the smaller truck fleets used for waste collection in constrained environments, such as narrow streets, by EVs. In this paper, we present a new formulation for the waste collection problem that considers a truck fleet comprised of smaller EVs and regular combustion trucks. The smaller trucks are proposed for the waste collection of specific sites (i.e. dumpsters in narrow streets). Our formulation considers battery limitations of electric trucks and flexible time windows for the waste collection task. The solution was validated by comparing the emission of CO2 and collection costs of a fleet comprised solely of combustion trucks and the hybrid fleet proposed here. The results showed that using a hybrid fleet significantly reduced waste collection costs and environmental impacts.