Abstract
This study explores the application of machine learning algorithms to optimize the geometry of the plasmonic layer in a surface plasmon resonance photonic crystal fiber sensor. By leveraging the simplicity of linear regression ( LR) alongside the advanced predictive capabilities of the gradient boosted regression (GBR) algorithm, the proposed approach enables accurate prediction and optimization of the plasmonic layer's configuration to achieve a desired spectral response. The integration of LR and GBR with computational simulations yielded impressive results, with an R-2 exceeding 0.97 across all analyzed variables. Moreover, the predictive accuracy demonstrated a remarkably low margin of error, epsilon < 10(-15). This combination of methods provides a robust and efficient pathway for optimizing sensor design, ensuring enhanced performance and reliability in practical applications.

Abstract

Abstract
Abstract High-performance computing (HPC) plays a crucial role in accelerating digital transformation, yet there is a lack of studies that systematically characterize its impact across different company sizes. This study addresses this gap by analyzing a cross-sectoral panel of 294 Portuguese companies, comprising 103 large enterprises and 191 small- and medium-sized enterprises (SMEs). It was applied descriptive analysis and statistical hypothesis testing methods. Two key research questions guide this investigation. The first explores the primary benefits and challenges associated with HPC adoption, while the second examines whether these factors vary between large companies and SMEs. The findings indicate that the benefits and challenges of the HPC are heterogeneously perceived by large companies and SMEs. It identified significant differences in the perceived benefits and challenges of HPC, particularly concerning cost savings, decision-making, cost and skills management, lack of awareness, and workforce skills gap. These findings contribute to a deeper understanding of how HPC supports digitalization processes, highlighting sector-specific and size-dependent differences in its perceived value and implementation barriers. This study provides valuable insights for businesses, policymakers, and researchers seeking to optimize HPC strategies for digital transformation.

Abstract
This study was initiated at a time of unprecedented uncertainty, as lecturers and educational institutions across the world tried to manage the move to online education as a result of the global COVID-19 pandemic. It started with lecturers' perspectives of their performance during that time to identify innovative teaching strategies beyond the priority of emergency teaching. The main goal was to identify the occurrence of more permanent changes in Higher Education after the pandemic. The research was based on a qualitative approach where faculty members were interviewed about their activities before, during and after lockdown periods. Data collected was analysed with the help of an algorithm based on Artificial Intelligence. Ultimately, it was possible to gather and evaluate practical solutions related to hybrid learning in Europe, Australia, and New Zealand, leading to recommendations for stakeholders in Higher Education.

Abstract
Globally Important Agricultural Heritage Systems (GIAHS) are models of sustainability, as they ensure a balance between human activity and ecosystem conservation. The Barroso region in Portugal is part of this network, as it follows traditional natural resource management and resilience practices by local communities. Given the threats posed by environmental degradation, it is urgent to adopt technological solutions for monitoring these conditions. Thus, throughout this article, the main threats to the integrity of these territories will be analyzed, and various methodologies and solutions for environmental monitoring will be presented. Based on the knowledge acquired, we will present an architecture for a digital solution that includes sensors, the Internet of Things (IoT), processing units, and platforms for real-time data visualization and alarm management.

Abstract
Text-to-SQL has gained increasing attention with Large Language Models (LLMs). While existing architectures have demonstrated the potential of multi-agent systems there remains significant room for improvement. In this work, we extend the CHESS framework by integrating In-Context Learning (ICL) techniques into the Candidate Generator module, evaluating three strategies: Zero-Shot, Few-Shot Learning, and Retrieval-Augmented Generation (RAG). We implement the system using GPT-4o, and perform experiments on the financial dataset from BIRD-SQL. Results show that Few-Shot Learning and RAG significantly outperform the standard approach. Compared to Zero-Shot (59.31% Execution Accuracy (EX), 0.412 ROUGE-1), RAG significantly boosted performance, increasing EX to 69.48% and ROUGE-1 to 0.652.