Abstract
Traditional machine learning techniques assume that data is drawn from a stationary source. This assumption is challenged in contexts with data streams for presenting constant and potentially infinite sequences whose distribution is prone to change over time. Based on these settings, detecting changes (a.k.a. concept drifts) is necessary to keep learning models up-to-date. Although state-of-the-art detection methods were designed to monitor the loss of predictive models, such monitoring falls short in many real-world scenarios where the true labels are not readily available. Therefore, there is increasing attention to unsupervised concept drift detection methods as approached in this paper. In this work, we present an unsupervised and interpretable method based on Radial Basis Function Networks (RBFN) and Markov Chains (MC), referred to as RMIDDM (Radial Markov Interpretable Drift Detection Method). In our method, RBF performs, in the intermediate layer, an activation process that implicitly produces groups of observations collected over time. Simultaneously, MC models the transitions between groups to support the detection of concept drifts, which happens when the active group changes and its probability exceeds a given threshold. A set of experiments with synthetic datasets and comparisons with state-of-the-art algorithms demonstrated that the proposed method can detect drifts at runtime in an efficient, interpretable, and independent way of labels, presenting competitive results and behavior. Additionally, to show its applicability in a real-world scenario, we analyzed new COVID-19 cases, deaths, and vaccinations to identify new waves as concept drifts and generate Markov models that allow understanding of their interaction.

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

Abstract
The growing number of senior experts leaving the workforce (especially in more developed economies, such as in Europe), combined with the ubiquitous access to artificial intelligence (AI), is triggering organizations to review their knowledge transfer programs, motivated by both financial and management perspectives. Our study aims to contribute to the field by analyzing options to integrate intergenerational tacit knowledge transfer (InterGenTacitKT) with AI-driven approaches, offering a novel perspective on sustainable Knowledge and Human Resource Management in organizations. We will do this by building on previous research and by extracting findings from 36 in-depth semi-structured interviews that provided success factors for junior/senior tandems (JuSeTs) as one notable format of tacit knowledge transfer. We also refer to the literature, in a grounded theory iterative process, analyzing current findings on the use of AI in tacit knowledge transfer and triangulating and critically synthesizing these sources of data. We suggest that adding AI into a tandem situation can facilitate collaboration and thus aid in knowledge transfer and trust-building. We posit that AI can offer strong complementary services for InterGenTacitKT by fostering the identified success factors for JuSeTs (clarity of roles, complementary skill sets, matching personalities, and trust), thus offering organizations a powerful means to enhance the effectiveness and sustainability of InterGenTacitKT that also strengthens employee productivity, satisfaction, and loyalty and overall organizational competitiveness.