Search Results (2,084)

This research advances the study of learning under non-stationary conditions by proposing a unified framework for concept drift detection and adaptive regression in evolving data streams. Unlike traditional batch models that assume static data distributions, the proposed approach operates sequentially, enabling real-time adaptation to drifting concepts in both time series and regression tasks. The method integrates Least Squares Support Vector Regression (LS-SVR) with Least Squares Support Vector Data Description (LS-SVDD) to jointly perform prediction and drift monitoring within a single kernel-based structure. LS-SVDD serves as a distributional drift detector, while LS-SVR incrementally updates model parameters to maintain predictive accuracy as data evolves. The framework accommodates both abrupt and gradual drifts, making it suitable for dynamic, high-dimensional environments. Experimental evaluations on synthetic data show that this proposal is able to outperform conventional batch and static methods in accuracy, responsiveness and computational efficiency. This method was compared using a real-world dataset, namely the high-dimensional Drosophila microarray time series, to demonstrate that the proposed approach is able to detect the meaningful change points using the whole data which is not doable using existing methods. Existing methods only used subsets of the dataset. These results highlight the potential of LS-SVR and LS-SVDD integration for real-time, adaptive learning across diverse domains where data distributions change over time. Full article

This paper addresses the problem that high pixel-level segmentation accuracy does not necessarily lead to geometrically compact building boundaries in vectorized outputs. A multi-task directional field learning framework is proposed based on U-Net with a ResNet-50 encoder. The framework introduces directional field supervision and a mask-field alignment loss to jointly optimize building region prediction and local boundary orientation consistency. In addition, a mild topological simplification procedure with a fixed small tolerance is applied to reduce residual staircase-like artifacts during vectorization. Experiments on the WHU building dataset at 0.2 m and 0.3 m spatial resolutions show that the proposed framework produces compact vector representations while maintaining high overlap relative to the raster reference annotations. In the 0.2 m setting, directional field learning improves Boundary IoU compared with the Baseline U-Net, whereas the complete pipeline slightly reduces Mask IoU and F1-score due to the additional simplification step. In the 0.3 m setting, the complete method does not consistently outperform several baselines in conventional pixel-level metrics, but it shows a favorable trade-off between polygon compactness and vector overlap under raster-reference evaluation. These results indicate that the proposed method is more suitable for geometry-aware vector reconstruction and vector simplification than for maximizing general semantic segmentation accuracy. In particular, the average number of polygon vertices is substantially reduced while Vector IoU remains approximately 90–92%. To further address the limitation of evaluating only on the WHU dataset, an additional in-domain validation experiment was conducted on the JAX dataset, which contains more complex building appearances and scene variations. The results show that the proposed Directional Field + Mild DP pipeline consistently reduces polygon complexity on the JAX dataset while maintaining competitive vector overlap. The central objective of the proposed framework is not only to improve mask-level building extraction, but also to enhance boundary-oriented vector reconstruction by learning local boundary-direction consistency and reducing raster-induced polygonal redundancy. Full article

As the modulation rate in high-speed optical communication systems continues to increase and modulation formats become increasingly complex, conventional electrical-domain sampling techniques, limited by the “electronic bottleneck,” are unable to meet the time-domain analysis requirements of optical vector signals with bandwidths exceeding 100 GHz. In this paper, a system based on linear optical sampling (LOS) is implemented for time-domain analysis of high-speed polarization-division-multiplexed (PDM) optical vector signals. An unbalanced input method is proposed to ensure the integrity of the sampling clock when the power of the signal under test is zero; a resampling method combined with soft integration is proposed to replace the conventional peak detection method, improving the accuracy of sampling point position and amplitude information extraction; and an adaptive frequency offset estimation algorithm is proposed to compensate for the continuously varying frequency offset caused by the use of low-repetition-rate sampling pulses. We constructed a signal acquisition system for optical vector signal measurement based on LOS. Using the above methods, the eye diagrams and constellation diagrams of 50 Gbaud PDM-QPSK (quadrature phase-shift keying), PDM-16QAM (quadrature amplitude modulation), and PDM-32QAM signals are successfully measured, and related parameters, including error vector magnitude (EVM) and signal-to-noise ratio (SNR), are calculated. The experimental results show that the proposed system achieves quasi-real-time measurement of 500 Gbps optical vector signals, and the measured performance parameters are on the same order of magnitude as those obtained from a commercial high-speed oscilloscope. Full article

Failures in critical systems and equipment within nuclear power plants (NPPs) significantly threaten operational safety and reliability. Therefore, rapid and accurate root cause localization during the incipient stages of failure is critical to preventing escalation. Traditional modeling methods often fail to address the inherent structural complexity of NPPs, the diversity of failure modes, and the stochastic mapping relationships between symptoms and causes. To address these challenges, this paper proposes an intelligent fault diagnosis framework integrating knowledge graphs (KGs) and Bayesian networks (BNs). First, by analyzing failure modes and anomaly characteristics, we define discrimination criteria for typical faults. Second, a structured knowledge modeling approach is developed to transform unstructured fault information into a KG, which is subsequently mapped to a BN topology. Finally, to mitigate the subjectivity of expert priors, data-driven structure and parameter learning algorithms are employed to optimize the model, enhancing inference accuracy. Robustness was validated through experiments targeting three fault severity levels, using signed directed graphs (SDGs), support vector machines (SVMs), domain generalization softmax (DG-softmax) and long short-term memory (LSTM) as benchmarks. Experimental results demonstrate that the proposed method maintains high diagnostic precision across varying severities, outperforming traditional data-driven methods in accuracy and stability. This study enhances the interpretability and engineering applicability of intelligent diagnosis in nuclear power systems. Full article

This paper demonstrates the application of a robust machine learning methodology to develop an accurate and, critically, an interpretable data-driven model for RC corbel shear assessment. A primary focus of this work is the use of advanced explainability techniques to rigorously validate the model’s predictive logic against fundamental principles of structural mechanics, directly confronting the limitations of “black-box” approaches. To implement this framework, an extensive database of 515 experimental tests was assembled. Different machine-learning (ML) techniques, including Random Forest, AdaBoost, Support Vector Machine, and XGBoost, were systematically evaluated to define the optimal predictive model. The most accurate algorithm, XGBoost, was selected and optimized to achieve exceptional performance, with a coefficient of determination ($R^2$) of 0.98 evaluated across the full database and a mean absolute relative deviation (MARD) of only 4%; on the held-out testing subset the model retains an $R^2$ of 0.97 and a MARD of 15%, confirming that predictive performance does not degrade appreciably on unseen specimens. The predictive model was shown to be substantially more accurate and generalizable than current design approaches, including both ACI code provisions and other prominent analytical models from the literature. Crucially, the Shapley Additive exPlanations (SHAP) technique was used to rigorously interrogate the model’s predictive logic. The analysis showed that the model’s feature attributions are consistent with established structural mechanics, correctly identifying the governing influence of parameters like the shear span-to-depth ratio and reinforcement indices for distinct failure modes. This explainability analysis establishes that the learned associations agree with structural expectations; it does not by itself demonstrate mechanistic causality. The study provides a validated methodology for creating trustworthy ML models and indicates, subject to further validation, uncertainty quantification, and a clearly defined applicability domain, how such interpretable tools might complement existing design provisions. Full article

Aligning industrial technological demands with academic expertise is critical for effective technology transfer. However, existing Expert Recommendation Systems (ERS) are frequently hindered by a “semantic gap” arising from terminological discrepancies between industry and academia, alongside a reliance on rigid classification taxonomies. To address these limitations, this paper proposes an automated expert finding framework that integrates Large Language Models (LLMs) with a hierarchical Retrieval-Augmented Generation (RAG) mechanism. Initially, we employ LLMs for the unsupervised extraction of research domains and technical keywords from heterogeneous multi-source data. To mitigate terminological diversity, we introduce a vector clustering-based Semantic Normalization module. By mapping diverse keyword variants into unified “Concept Clusters,” this module reduces vocabulary sparsity by 98%. These organized clusters are structured into a “Semantic Tree” to support a hierarchical RAG strategy, enabling a coarse-to-fine retrieval process from broad disciplinary domains down to specific technical achievements. In this paper, RAG refers to a retrieval-augmented expert recommendation workflow, in which the retrieved achievements and expert evidence are used as grounded context for generating an explanatory recommendation report. Evaluations on a real-world dataset show that the framework achieves a Precision@5 of 78.4% and a Recall@10 of 81.2%, outperforming flat vector retrieval baselines by over 20% in precision. Furthermore, hierarchical domain pruning significantly reduces computational overhead, decreasing average query latency by a factor of three (to 115.8 ms). These results demonstrate that the proposed system effectively bridges the industry–academia semantic gap, providing a scalable and accurate solution for expert recommendation. Full article

Microplastics (MPs, plastic particles < 5 mm) and nanoplastics (NPs, plastic particles generally <1 µm), collectively referred to as micro/nanoplastics (MNPs), have emerged as critical contaminants in wastewater systems due to their persistence, small size, and ability to act as vectors for co-contaminants. Conventional wastewater treatment technologies are often insufficient for the effective removal of microplastics, particularly for smaller particles and nanoplastics, necessitating the development of functional materials and innovative treatment strategies. In this review, recent advances in carbon-based materials, cellulose-based materials, and their hybrid carbon–cellulose composites for microplastics removal are critically analyzed and comparatively discussed. Particular attention is given to the structure–function relationships governing adsorption performance, including the roles of hierarchical porosity, surface chemistry, and interfacial interactions. The key mechanisms responsible for microplastics capture—such as hydrophobic interactions, π–π stacking, hydrogen bonding, electrostatic attraction, physical entrapment, and pore trapping—are systematically discussed. Carbon–cellulose composite materials are highlighted as a promising class of multifunctional adsorbents due to their synergistic combination of hydrophilic cellulose scaffolds and hydrophobic carbon domains. This dual functionality enables efficient removal of microplastics across a wide range of sizes and morphologies. Recent developments in magnetic and superhydrophobic composite systems further demonstrate enhanced separation efficiency, recyclability, and potential applicability in real wastewater environments. In addition to summarizing recent progress, this review critically examines the methodological inconsistencies, mechanistic uncertainties, and practical limitations associated with current adsorption systems. Despite significant progress, several challenges remain, including the lack of standardized evaluation methods, limited validation under real wastewater conditions, material stability issues, and scalability constraints. Future research directions are proposed, focusing on rational material design, sustainable carbon sources, multifunctional hybrid systems, and integration into existing treatment infrastructures. The development of sustainable hybrid adsorption systems for microplastics remediation also contributes to the achievement of Sustainable Development Goal 6 (Clean Water and Sanitation) by supporting improved wastewater treatment technologies and reduction in emerging aquatic contaminants. Full article

While social media platforms are primary vectors for misinformation, automated detection systems remain largely confined to English. This paper presents a transferable, three-stage framework for fine-tuning transformer models to detect domain-specific deceptive content in Spanish. The pipeline comprises: (1) corpus unification, merging fragmented datasets into a 61,674-article resource mapped into three classes (Real, Fake, Satire) to prevent stylistic confounding; (2) systematic model optimization, extensively benchmarking classical metaheuristics against eight transformer architectures (including mBERT, XLM-RoBERTa, and BETO) using strong regularization to mitigate overfitting; and (3) production deployment, encapsulating the optimized model as a containerized web application for real-time inference. Through rigorous experimentation, the Spanish-specific BETO encoder emerged as the strongest model for this task, achieving 89.18% overall accuracy. The model attains a near-perfect in-source F1-score on the satire class; however, a strict source-held-out test reveals that this performance is highly source-dependent—recall on satire from an unseen outlet drops to 0.08—indicating that single-source class construction leads the model to recognize the source rather than a generalizable category. We report this finding as a central methodological result: corpus design, and in particular the source diversity of each class, is the primary determinant of whether the framework generalizes. Adversarial robustness tests using named-entity masking and typo injection provide complementary evidence on the model’s reliance on semantic versus surface cues. The methodology is designed to be adaptable across domains: by substituting the training corpus, the same framework may in principle be retargeted to other digital threats, such as investment scams and phishing, provided that suitable labeled corpora are constructed and validated for each new domain. The complete framework, dataset, and application are released as open-source resources to support reproducible research and practical countermeasures against online misinformation. Full article

Assessing motor symptoms in Parkinson’s disease (PD) is challenging due to the progressive evolution of the condition and the variability of symptoms, which are not fully captured by periodic clinical visits. In this context, wearable sensors and machine learning (ML) have emerged as a viable path toward objective and continuous monitoring, although achieving robust generalization to free-living conditions remains a challenge. This work explores the egg-beating task, a simple everyday activity, as a digital approach for PD motor assessment using smartwatch-based inertial measurements and ML techniques. Twenty-two individuals with PD and sixteen healthy controls (HC) completed a one-minute egg-beating task while wearing a smartwatch equipped with tri-axial accelerometer and gyroscope sensors. Data were recorded both under supervised clinical conditions and during unsupervised home sessions. Time- and frequency-domain features were extracted from the inertial signals, and models trained exclusively on supervised recordings were then tested on supervised, unsupervised, and combined data. PD participants showed systematically lower movement amplitude, slower oscillation frequency, and a progressive drop in signal energy over the course of the task, all of which align with the characteristic features of bradykinesia. The support vector machine achieved the best overall performance, reaching 90% accuracy in distinguishing PD from healthy controls under supervised conditions, with a reduction of less than 4% when applied to unsupervised data. These results support the egg-beating task as a practical and ecologically valid method for real-world motor assessment, with potential for future use in remote monitoring and longitudinal assessment. Full article

Accurate viral protein function annotation is essential for genomic surveillance, yet conventional retrieval-augmented generation (RAG) pipelines often fragment biological evidence into fixed-length text chunks, disrupting relationships among ORFs, annotations, structural domains, sequence motifs, residue mappings, and model-derived attention evidence. We propose ViroBioTree, a tree-structured biological evidence retrieval framework for downstream viral protein evidence review rather than a new primary annotation classifier. Built as an evidence organization layer on ViralMultiNet-derived ORF-level predictions and annotations, ViroBioTree converts sequence, annotation, structure, and attention evidence into typed biological nodes and traceable edges, then performs deterministic multi-channel recall, evidence-aware reranking, balanced TopK selection, rule-based verification, and node-cited report generation. In a demo benchmark, ViroBioTree achieved its strongest deterministic proxy performance on structure-explanation tasks, with Precision@K = 1.0, Recall@K = 1.0, and diversity = 0.52; these values reflect expected node-type and tag agreement rather than independent biological correctness. A bounded full-scale SARS-CoV-2 index contained 39,800 ORF rows, 80,000 attention records, 199,418 nodes, and 495,886 edges. In a stratified full20k diagnostic evaluation, ViroBioTree showed task-dependent advantages over LlamaIndex vector retrieval for conflict detection, evidence retrieval, and structure explanation, while LlamaIndex remained competitive or stronger for annotation-rich function annotation. A cross-family Influenza A Virus (IAV) diagnostic audit showed that the schema can represent IAV evidence namespaces while explicitly exposing missing formal ORF inputs, missing attention evidence, and unavailable residue/PDB assertions. Supplementary robustness, external sanity-check, diversity-risk, expert-evaluation, domain-tool positioning, and cross-family audit analyses supported traceability, report quality, and conservative evidence handling, but also showed that stable Precision@K under query perturbation does not necessarily imply stable retrieved evidence sets. ViroBioTree operates offline and deterministically, but does not address raw-read assembly, base calling, primary ORF prediction, or wet-lab validation. Its results should be interpreted as proxy and expert-reviewed evidence for traceable viral protein evidence retrieval and report generation rather than as direct validation of biological function annotation. Full article

The exponential growth of cyber threats in modern digital infrastructure demands advanced detection systems that adapt to evolving attack patterns. Traditional cybersecurity approaches struggle with dynamic threats, requiring extensive labeled datasets and retraining for each new category. This paper presents a comprehensive transfer learning framework for cybersecurity threat detection, leveraging the CICIoMT dataset as a benchmark to enhance detection capabilities across heterogeneous cybersecurity environments. We propose a machine learning (ML)-enabled framework that employs systematic feature alignment, hybrid class balancing, and multi-algorithm evaluation using machine learning models, including Multi-Layer Perceptron (MLP), Support Vector Machine (SVM), Random Forest (RF), Gradient Boosting, and XGBoost. The proposed approach addresses the critical challenges of data scarcity and domain heterogeneity in cybersecurity by enhancing feature engineering with cybersecurity-specific features, statistical aggregations, and PCA embeddings. Extensive experimental evaluation across two target datasets (CICIoT and IoT-23) demonstrates both the exceptional successes and critical limitations of cross-domain transfer learning in cybersecurity. The framework achieved outstanding performance on domain-compatible datasets, with RF reaching 99.0% accuracy on CICIoT, Gradient Boosting achieving 98.9%, and XGBoost delivering 98.4%, demonstrating exceptional knowledge transfer from medical IoT to smart home IoT environments. However, transfer learning to IoT-23 was unsuccessful (50% accuracy, equivalent to random guessing), revealing that feature domain difference, where identical attack labels encode fundamentally different behavioral patterns, prevents effective knowledge transfer despite nominal class overlap. This research makes significant advances in adaptive cybersecurity systems by providing a rigorous evaluation of both the successes and limitations of transfer learning. This work demonstrates that ensemble methods (RF, XGBoost, and Gradient Boosting) achieve superior cross-domain performance compared with neural networks on compatible domains, while also revealing fundamental challenges when the source and target domains differ in their feature spaces. Full article

Accurate prediction of student success is important for improving retention and supporting academic decision-making. This study investigates the use of Graph Neural Networks (GNNs) for modelling academic progression across 13 university faculties, focusing on how structured trajectory representations can capture temporal and relational dependencies in student records. Conventional machine learning approaches frequently rely on aggregated tabular indicators, which may obscure the temporal ordering and relational structure of academic trajectories. We propose a graph-based representation, termed the Academic Trajectory Graph (ATG), which encodes semester-to-semester course transitions and temporal relationships between enrollments. To evaluate its effectiveness, graph-based approaches (DGCNN, GCN, and node2vec-based embeddings) are compared with conventional machine learning models trained on aggregated tabular indicators. Experimental results show that aggregated academic indicators provide strong predictive performance, confirming the importance of cumulative progression signals. Among the graph-based approaches, DGCNN achieved the most consistent performance across faculties, reaching AUC values up to 0.988, accuracy up to 0.945, and F1-scores above 0.94 in several faculties. Although traditional tabular models generally achieved the strongest overall predictive performance, the ATG preserves temporal and relational information that is not captured by flat feature vectors, enabling trajectory-level analysis of academic pathways. The results indicate that graph-based trajectory modelling may provide additional structural insights in faculties exhibiting heterogeneous academic progression patterns, while offering competitive predictive performance. These findings highlight both the potential and the limitations of graph-based approaches for modelling structured educational data, suggesting that their effectiveness depends on the structural diversity of the underlying academic domain. Full article

Paper-based logistics documents remain widely used in multi-enterprise supply chains, where heterogeneous layouts, noisy document images, non-standard material descriptions, and limited edge-computing resources make structured extraction and material matching difficult. This paper proposes RRA-Logis, a lightweight multimodal large-language-model framework for logistics document understanding and material entity alignment. Instead of treating logistics document processing as a conventional field-extraction task, RRA-Logis formulates it as a document-to-entity alignment problem under resource constraints. The framework combines schema-constrained image-to-JSON extraction, LoRA/QLoRA instruction tuning, vector-based candidate recall, LLM-based semantic verification, confidence-gated decision making, and human-in-the-loop data evolution. Its methodological contribution lies in organizing these components into a resource-aware decision mechanism that determines whether a material-matching result should be automatically accepted or routed to human verification according to confidence and ambiguity margins. Experiments under a 24 GB VRAM constraint show that the fine-tuned Qwen2.5-VL-7B model achieves 85.4% document-level extraction accuracy and 100% JSON compliance on L-Doc-2K, while the proposed two-stage material-alignment method achieves 92.8% Top-1 accuracy. Ablation results indicate that LLM-based re-ranking, fused scoring, and confidence-gated verification each contribute to improved alignment reliability. Additional evaluation on the public DocILE benchmark and a desensitized real-document subset further examines cross-domain extraction transfer and the gap between Sim-to-Real data and operational logistics documents. The results suggest that RRA-Logis provides a practical framework for logistics document automation under constrained computing resources, while larger-scale real-world validation and broader benchmarking against specialized document-intelligence systems remain necessary. Full article

Single-domain antibodies (sdAbs) are the smallest antigen-binding antibody (Ab) fragments (12–15 kDa) and have emerged as a versatile therapeutic platform. Their compact size, high solubility, stability, and ability to access cryptic epitopes distinguish them from conventional monoclonal Abs (mAbs) and larger Ab fragments. These properties are particularly attractive in ophthalmology, where molecular size, tissue penetration, and formulation constraints critically influence therapeutic performance. This narrative review summarizes the structural features, engineering strategies, immunogenicity considerations, and production platforms of sdAbs, with a focus on ocular applications. Preclinical studies demonstrate promising efficacy in retinal vascular diseases through targeting of VEGFA, ANG2, TNFα, and complement components, as well as in inflammatory and anterior segment disorders. SdAbs can be formatted as multimeric or Fc-fused constructs to extend intraocular half-life or delivered via gene therapy vectors as a sustained intraocular “biofactory” approach. Notably, recent work demonstrates the feasibility of vector-encoded sdAbs targeting complement C3 in vivo. While challenges remain regarding immunogenicity, pharmacokinetics, and regulatory pathways, the approval of several sdAb-based drugs in other fields underscores their clinical potential. SdAbs represent a promising next-generation modality for ocular therapeutics, enabling innovative strategies beyond conventional antibody formats. Full article

This study presents a unified computational framework for detecting chaotic behavior in a fractional-order Lorenz system by combining Lyapunov-based dynamical analysis with modern machine learning methods. The fractional-order system is simulated over a wide range of the control parameter, and the corresponding Lyapunov spectrum is computed to identify chaotic and non-chaotic regimes. These time-domain trajectories are transformed into high-resolution wavelet scalogram images, enabling a vision-based representation of fractional-order dynamics. The resulting image dataset is classified using both a Vision Transformer (ViT) model and a Support Vector Machine (SVM) classifier built on ViT-extracted feature embeddings. Experimental results demonstrate that the ViT model achieves near-perfect discrimination between chaotic and non-chaotic patterns, with an accuracy of 0.9627, a Cohen’s kappa of 0.920, and an MCC of 0.949. The SVM classifier yields even higher performance, achieving an accuracy of 0.9776, a kappa coefficient of 0.955, and an MCC of 0.955. ROC analyses confirm that both models reach an AUC of 1.00, indicating excellent separability between the two classes. The findings show that wavelet-based image encoding combined with transformer architectures provides a powerful and generalizable approach for chaos detection in fractional-order nonlinear systems. This integrated methodology offers a scalable solution for automated analysis of complex dynamical behavior and establishes a bridge between classical chaos theory and state-of-the-art deep learning models. Full article