Search Results (7,707)

Financial time series forecasting remains a challenging task due to the high non-stationarity and concept drift inherent to market data. Existing deep learning models, such as LSTMs and transformers, typically employ static weights after training, limiting their ability to adapt to rapid market regime shifts (e.g., from trends to reversals). To bridge this gap between static parameters and dynamic environments, we propose a novel framework named Game-Theoretic HyperNetwork (GTH-Net), which introduces a context-aware meta-learning mechanism to achieve adaptive forecasting. Specifically, we first introduce an Evolutionary Game-Theoretic Correction Module (E-GTCM) to explicitly extract latent buying and selling pressure based on market microstructure priors through an iterative gated evolution process. Subsequently, we propose a HyperNetwork-based fusion mechanism that treats the extracted game state as a meta-context to dynamically generate the weights of the forecasting head. This allows the model to automatically switch its prediction rules in response to shifting market regimes. Extensive experiments on real-world stock datasets demonstrate that GTH-Net significantly outperforms baselines in terms of machine learning predictive accuracy and simulated financial profitability. Furthermore, ablation studies and parameter analysis confirm that the dynamic weight generation mechanism effectively captures market reversals caused by overcrowded trades. Full article

Adenosine triphosphate (ATP) serves as a central energy currency and signaling molecule in cellular processes, with ATP-binding sites in proteins playing critical roles in enzymatic catalysis, signal transduction, and gene regulation. The accurate identification of ATP-binding sites is essential for understanding protein function mechanisms and facilitating drug discovery, enzyme engineering, and disease pathway analysis. In this study, we present a novel hybrid deep learning framework that synergizes heterogeneous learning paradigms based on protein sequence information for accurate ATP-binding site prediction. Our approach integrates two complementary base classifiers. One is a Transformer-based model, which leverages high-level contextual embeddings generated by Evolutionary Scale Modeling 2 (ESM-2), a state-of-the-art protein language model, combined with a local–global dual-attention mechanism that enables the model to simultaneously characterize short-segment and long-range contextual dependencies across the entire protein sequence. The other is a deep Q-network (DQN)-inspired classifier that achieves residue-level prediction as a sequential decision-making process. The final predictions are generated using a weighted ensemble strategy, where optimal weights are determined via cross-validations to leverage the strengths of both models. The prediction results on benchmark independent testing sets indicate that our method achieves satisfactory performance on key metrics. Beyond predictive efficacy, this work uncovers the intrinsic biological mechanisms underlying protein–ATP interactions, including the synergistic roles of local structural motifs and global conformational constraints, as well as family-specific binding patterns, endowing the research with substantial biological significance. The research in this work offers a deeper understanding of the protein–ligand recognition mechanisms and supportive efforts on large-scale functional annotations that are critical for system biology and drug target discovery. Full article

Imine reductases (IREDs) have emerged as valuable biocatalysts for the asymmetric synthesis of chiral amines, key intermediates in numerous active pharmaceutical ingredients. Their ability to operate under mild reaction conditions with high chemo- and stereoselectivity provides an attractive alternative to conventional metal-catalyzed or chemical reduction processes. However, the broader industrial application of wild-type IREDs is often constrained by their limited substrate scope and moderate catalytic efficiency. Recent advances in biocatalysis have demonstrated that engineered IREDs can catalyze the reduction of a wide range of natural and non-natural imines, significantly expanding their applicability in pharmaceutical and fine chemical synthesis. In parallel, enzyme immobilization strategies have proven highly effective for improving operational stability, facilitating enzyme reuse, and enabling continuous flow biocatalytic processes. Efficient cofactor regeneration systems have further enhanced the practical implementation of IRED-based transformations. Advances in protein engineering, including structure-guided design, semi-rational mutagenesis, and directed evolution, have generated enzyme variants with improved catalytic activity, stereoselectivity, and substrate tolerance. The integration of high-throughput screening technologies and machine-learning-assisted enzyme design has further accelerated the discovery and optimization of efficient IRED biocatalysts. This review summarizes recent progress in the protein engineering and immobilization of IREDs and discusses future perspectives for their industrial application. Full article

Cellulose paper, a natural polymeric dielectric, determines the lifetime of oil–paper insulation systems in transformers, yet its molecular degradation behavior in ester-based insulating media remains insufficiently clarified. This study investigates the electro–thermal aging of cellulose polymer immersed in soybean-based natural ester (SBNE) and palm fatty acid ester (PFAE), with emphasis on depolymerization and its relationship with partial discharge (PD) activity. Accelerated aging experiments were conducted under combined electrical and thermal stress, and the evolution of the degree of polymerization (DP) was measured to quantify polymer chain scission. Phase-resolved PD (PRPD) patterns were recorded during aging, and multi-dimensional statistical features were extracted and reduced using principal component analysis to characterize degradation-sensitive electrical responses. The results show a progressive decrease in DP with aging time in both ester media, accompanied by distinct PD evolution characteristics, indicating different influences of the two esters on cellulose polymer stability. An ensemble learning model integrating multiple classifiers was further employed to identify aging stages based on PD features, achieving reliable discrimination performance. These findings establish a correlation between cellulose depolymerization and dielectric discharge behavior, providing a polymer-centered interpretation of aging mechanisms in ester-based oil–paper insulation systems. Full article

Short-term forecasting of solar radiation is essential for the efficient operation of solar energy systems. This study presents a neural network-based approach for short-term solar radiation forecasting using a hybrid framework that integrates temporal characteristics with weather-based features. The proposed model combines a Gated Recurrent Unit (GRU) to capture short-term temporal dynamics, a Transformer Encoder, and a Multilayer Perceptron (MLP) to integrate these representations for final prediction. Key meteorological variables, including temperature, humidity, and wind speed, are incorporated along with engineered time-related features such as lagged values, rolling statistics, and cyclical time-of-day encodings. The results demonstrate that the hybrid model effectively integrates sequential learning and feature interaction, leading to improved forecasting accuracy. The proposed approach achieves a test Mean Absolute Error (MAE) of 0.056, Root Mean Square Error (RMSE) of 0.086, and coefficient of determination (R²) of 0.92, outperforming benchmark models such as AutoRegressive Integrated Moving Average (ARIMA), Long Short-Term Memory (LSTM), GRU, and Extreme Gradient Boosting (XGBoost). The model maintains stable performance across cross-validation folds, multiple forecasting horizons, and varying weather conditions. These findings indicate that the proposed framework provides a reliable and practical solution for accurate short-term solar radiation forecasting, supporting real-time solar energy management and renewable energy system optimization. Full article

The school garden has a long-standing pedagogical tradition linked to active, experiential, and community-based education, represented by authors such as Montessori, Freinet, and Dewey. Currently, its role has been consolidated as a relevant educational resource used to address the challenges of sustainability education, pedagogical innovation, and student holistic development. This research takes a qualitative approach based on a multiple case study conducted in four educational centers in Spain and Portugal. Semi-structured interviews, documentary analysis, and reflective memoranda were used. Content analysis was performed using a deductive–inductive coding approach in ATLAS.ti software v. 25th, combining literature-derived categories with those emerging from the data, following a thematic analysis (TA) approach. The results suggest that school gardens promote meaningful learning, the development of transversal competencies, improved school climate, and community involvement. Pedagogical, social, and emotional benefits were identified, as well as high levels of satisfaction among all participants. However, obstacles were found to persist, mainly related to a lack of time and teacher coordination. The study confirms that the school garden serves as a pedagogical resource with a high transformative potential. Its effectiveness depends on intentional curricular integration, teacher commitment, and the engagement of the educational community, aligning with the principles of an active, sustainable, and contextualized pedagogy. Full article

Early detection of battery degradation is essential for ensuring the safety and reliability of electric vehicle (EV) systems under real-world operating variability. This paper proposes a physics-guided multi-sensor learning framework, termed SensorFusion-Former (SFF), for early warning of short-term EV battery performance degradation. The proposed approach integrates a physics-based baseline model for operational normalization, a multi-sensor fusion attention mechanism to model cross-modality interactions, and a lightweight transformer architecture for efficient temporal representation learning. Weak supervision is derived from physics-consistent residual analysis with temporal smoothing, enabling scalable training without dense manual annotations. To support reliable deployment, evidential uncertainty modeling and conformal calibration are incorporated to obtain statistically controlled decision thresholds. Experiments conducted on a real driving cycle dataset from IEEE DataPort demonstrate that SFF consistently outperforms classical machine learning methods, deep neural networks, and standard transformer models in terms of early-warning lead time, false alarm rate, and inference efficiency while maintaining competitive discriminative performance. Cross-scenario evaluations under diverse thermal conditions further confirm the robustness and generalization capability of the proposed framework. Full article

Over the past decade, road safety research has experienced rapid development due to the rapid expansion of large crash databases, the adoption of artificial intelligence techniques, and the demand for proactive and predictive safety solutions. This study conducts a data-driven review of recent research trends in transport safety. It focuses on main domains including crash severity analysis, human factors, vulnerable road users (VRUs), spatial modeling, and artificial intelligence applications. A systematic search of the Scopus database identified 15,599 relevant scientific papers published between 2016 and 2025. After constructing this corpus, titles, abstracts, and keywords were preprocessed using a natural language pipeline. The analysis employed BERTopic, a transformer-based topic modeling framework. The analysis identified 29 distinct research topics, further synthesized into five major thematic areas: (1) crash severity and injury analysis, (2) driver behavior and human factors, (3) vulnerable road users, (4) artificial intelligence, machine learning, and computer vision in intelligent transportation systems, and (5) spatial analysis and hotspot detection. A notable increase in publications related to artificial intelligence and machine learning has been evident since 2020. The results show a transition from descriptive, post-crash studies to integrated, multimodal, predictive analysis. Overall, the findings reveal a paradigm shift in the field. This study also identifies ethical and economic issues associated with the use of artificial intelligence in intelligent transportation systems, including data management, infrastructure requirements, system security, and model transparency. The results signify a transition from intuition-based models to explainable, spatially explicit, and data-intensive models, ultimately facilitating proactive risk assessment and informed decision-making. Full article

Addressing the challenge of coordinated dispatch between wind/solar and thermal power in new energy grids, this research proposes a thermal power unit output prediction method based on a Transformer-BiLSTM hybrid deep learning model. First, a simulated annealing algorithm optimizes the output configuration of solar thermal power plants to mitigate fluctuations in wind and solar combined generation. An ant colony-greedy algorithm is then integrated to determine the optimal dispatch data for thermal power units, constructing a high-quality training dataset under physical constraints. In the model design, a bidirectional long short-term memory network captures short-term temporal features, while the Transformer’s multi-head self-attention mechanism models long-term dependencies. The model innovatively incorporates the learnable positional encoding to enhance temporal awareness. Experimental results demonstrate accurate predictions, with the power constraint mechanism effectively correcting over-limit forecasts. This ensures 98.7% of predictions during low-load periods comply with unit technical specifications. Compared to existing methods, this model avoids data limitations and manual feature engineering bottlenecks through the end-to-end wind–solar–thermal mapping, providing a high-precision solution for dispatch decisions in renewable-dominated grids. Full article

This paper aims to present a non-destructive, optimized variational mode decomposition (VMD)-based ground-penetrating radar (GPR) method developed for identifying void defects in reinforced concrete (RC) structures. This study also presents an enhanced framework for defect detection in RC by integrating advanced spectrum analysis with deep learning techniques. A GPR investigation was conducted on an RC bridge deck with known structural defects to generate a representative dataset reflecting both intact and void-defective conditions. In addition to conventional spectral techniques such as fast Fourier transform (FFT), spectrogram, and scalogram, an optimized variational mode decomposition (VMD) method was implemented. The VMD approach decomposes GPR signals into intrinsic mode functions, enabling refined feature extraction beyond traditional spectral methods and allowing clear differentiation between intact and defective signals. The limited availability and quality of GPR small datasets have restricted the application of a functional 1D-CNN which generally requires at least several hundred datasets. To address this challenge, a data augmentation strategy is adopted. FFT-based features were successfully utilized to train a one-dimensional convolutional neural network (1D-CNN) for automated defect identification. The results demonstrate that both the advanced spectrum-based approach and the hybrid framework combining spectral analysis with deep learning significantly improve defect detection performance. Overall, the proposed methodology provides an effective and intelligent solution to support timely, data-driven decision-making for maintenance and safety assurance of bridge infrastructure. Full article

The rapid expansion of mobile healthcare (mHealth) applications has transformed health-related e-commerce, creating new challenges for understanding and responding to user needs. This study proposes a data-driven framework to systematically identify and prioritize unmet user requirements from negative reviews of Chinese mHealth applications. Using a dataset of 31,124 user reviews collected between 2019 and 2025, the framework integrates sentiment analysis, topic modeling, and machine learning regression to uncover six key areas of user concern and examine their temporal evolution. Among several predictive models linking user concerns to app ratings, the k-nearest neighbors (KNN) model demonstrated superior performance. Subsequent SHAP-based interpretability analysis reveals that account authentication, system accessibility, and application stability have the most significant impact on user ratings, highlighting the critical roles of trust and technical reliability in health e-commerce. This research not only provides actionable insights for platform governance but also contributes a generalizable methodology for leveraging user-generated content to inform evidence-based management and policy decisions in mobile digital services. Full article

Accurate assessment of hypoxia-related fetal risk during labour is essential for improving perinatal outcomes while avoiding unnecessary operative interventions. Although deep learning has shown promise for automated fetal risk assessment, most existing approaches rely on cardiotocography (CTG) alone; CTG interpretation is known to suffer from a high false-positive rate and may not fully reflect fetal status without complementary clinical context. To address this limitation, we propose MIRF-Net, a multimodal intrapartum fetal risk assessment framework that jointly models (i) CTG time-series signals, (ii) Gramian Angular Difference Field (GADF) images that encode global correlation structure of fetal heart rate, and (iii) structured maternal metadata. MIRF-Net combines a PatchTST encoder for CTG, a pretrained ResNet101 for GADF images, and an autoencoder for maternal metadata and then performs cross-modal interaction learning with a fusion Transformer for final risk prediction. Using 552 eligible CTG recordings from the public CTU-UHB intrapartum database, which were split into training, validation, and test sets at a ratio of 6:2:2, MIRF-Net outperforms representative baselines on the test set, achieving a quality index (QI) of 74.76%, AUC of 0.7413, and Brier score of 0.2537, indicating improved discrimination and better-calibrated risk probabilities. Ablation studies further confirm the complementary contributions of each modality and show that Transformer-based fusion yields the most consistent overall gains. These results suggest that MIRF-Net provides reliable decision support for intelligent intrapartum monitoring. Full article

In abnormal goat lung sound recognition, high inter-class similarity and large intra-class variability pose significant challenges. To address this issue and improve recognition performance, we propose a deep learning model, AAF-SwinT, based on an improved Swin Transformer. The model replaces the original Swin Transformer self-attention module with Axial Decomposed Attention (ADA), modeling the temporal and frequency axes separately and integrating attention weights to mitigate inter-class feature similarity. Adaptive Spatial Aggregation for Patch Merging (ASAP) is designed to emphasize key time-frequency regions, and a Frequency-Aware Multi-Layer Perceptron (FAM) is introduced to model features across different frequency bands, further enhancing the discriminative ability for abnormal lung sounds. Experiments on a self-constructed goat lung sound dataset demonstrate that AAF-SwinT achieves an accuracy of 88.21%, outperforming existing mainstream Transformer-based models by 2.68–5.98%. Ablation studies further confirm the effectiveness of each proposed module, improving the accuracy of baseline Swin Transformer model from 85.53% to 88.21%. These results indicate that the proposed approach exhibits strong robustness and practical potential for abnormal lung sound recognition in goats, providing technical support for early diagnosis and management of respiratory diseases in large-scale goat farming. Full article

Background. Smart warehousing increasingly relies on digital twin technologies to enhance operational efficiency, real-time visibility, and decision-making in logistics systems. However, existing research primarily focuses on technological capabilities while paying limited attention to the organizational practices that shape successful implementation. Methods. This study aims to identify and prioritize the critical success factors (CSFs) for integrating digital twins into smart warehousing using the Practice-Based View (PBV) as the theoretical lens. Based on insights from prior research and expert validation, nine CSFs were identified and evaluated using the Best–Worst Method (BWM). Empirical input was obtained from six industry experts with experience in digital transformation, warehousing, and supply chain management. Results. The results indicate that collaborative learning, contextual training, and gamification elements emerge as the most influential critical success factors, highlighting the importance of organizational practices in supporting digital twin adoption in smart warehousing. Conclusions. By linking technological capabilities with organizational routines, the proposed framework provides both theoretical insights and practical guidance for implementing digital twins in smart warehouse environments. Full article