Search Results (405)

The Leaf Area Index (LAI) serves as a vital biophysical parameter for quantifying vegetation dynamics and ecosystem functioning. While traditional LAI retrieval methods face challenges in handling nonlinear spectral-vegetation relationships, machine learning (ML) approaches offer promising alternatives through their data-driven adaptability. This study presents a comprehensive cross-site assessment of 13 ML algorithms for LAI estimation, leveraging ground observations from 98 sites worldwide. Our systematic assessment reveals three key findings: First, ensemble methods consistently outperformed other approaches, with Gradient Boosted Tree Regression (GBTR) achieving superior accuracy (R² = 0.647, RMSE = 0.899) and robustness (ΔR² < 0.05 beyond n = 69 training samples). Second, Gaussian Process Regression (GPR) illustrated exceptional stability across varying training sizes (R² = 0.607 ± 0.012), highlighting its reliability for data-limited scenarios. Third, all tested ML models substantially outperformed operational LAI products, with the GBTR model demonstrating superior explanatory power (external validation R² = 0.647) compared to MODIS; its R² value had increased by 0.489. This optimal balance of accuracy, computational efficiency, and resistance to overfitting positions GBTR as a reasonable choice for large-scale LAI mapping. These findings underscore ML’s promising potential in vegetation monitoring while highlighting the need for hybrid approaches that combine physical principles with data-driven learning to address current limitations in extreme-value estimation and ecological generalizability. Full article

Airport systems rely on tightly connected digital and physical components, so cyber disruptions can affect both service performance and passenger movement. Existing airport simulation studies often focus on either queue-based passenger processing or pedestrian movement but rarely combine both in a framework suited for cyber-resilience analysis. This paper presents a hybrid simulation framework that integrates discrete-event simulation (DES), JuPedSim-based microscopic pedestrian modeling, and structured large language model (LLM) decision support to examine how cyber disruptions propagate through passenger-facing airport operations. The DES layer models service processes such as check-in, information desks, and security screening, while the pedestrian layer models movement, congestion, route choice, and spatial occupancy. Under degraded display or guidance conditions, the LLM generates structured passenger-level post-security decisions, such as going directly to the gate, checking a display, asking staff, waiting, visiting optional activity areas, or first moving to a wrong intermediate area. The framework is evaluated through a 500-passenger terminal case study with one baseline case and four disruption cases. Results show that check-in and security degradation produce the largest throughput loss, queue growth, and completion-time increase, while guidance degradation mainly affects post-security behavior. Spatial heatmaps further show where bottlenecks emerge and how congestion shifts across the terminal. Additional Rotterdam checkpoint validation, Palma benchmark analysis, and LLM ablation results support the framework’s ability to reproduce plausible queue, timing, throughput, and behavior-sensitive disruption patterns. The study provides a practical methodology for exploratory airport cyber-resilience assessment under coupled service, movement, and degraded-guidance conditions. Full article

The transition to a hydrogen-based energy economy emphasizes the potential of hydrogen as a fuel for plug-in hybrid electric vehicles (PHEVs). The performance of a hydrogen engine within a PHEV depends on the choice of its operating modes, which influence both efficiency and emissions. This study proposes a method for developing engine operating lines (EOLs) on engine maps based on minimizing nitrogen oxide (NOx) emissions while considering constraints on maximum engine power. A total of 15 EOLs are proposed for configurations with both constant and variable maximum engine power. Using mathematical modeling of PHEV operation under the Worldwide Harmonized Light Vehicles Test Cycle (WLTC), the impact of EOL selection on engine characteristics, as well as on battery and generator parameters, is analyzed. For a comprehensive evaluation of EOL effectiveness, five criteria are introduced, considering fuel energy consumption, NOx emissions, wear, mechanical fatigue, and noise, vibration, and harshness (NVH). The Analytic Hierarchy Process (AHP) and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) are applied to determine the weighting factors of the criteria and to rank the proposed EOLs, thereby identifying the most efficient configurations. The results show that, for the base hydrogen engine configuration, selecting appropriate operating modes alone enables NOx emissions to be reduced significantly below Euro 6 limits, without any hardware modifications or exhaust aftertreatment. Full article

This article examines remittances not only as financial transfers but also as datafied political objects shaped by measurement, modelling and presentation infrastructures. Using the UK–Romania corridor, we compare observed personal remittance receipts published by the National Bank of Romania (NBR) with model-based bilateral estimates associated with World Bank/KNOMAD data. The article develops an analytical framework that links quantification, metric power, algorithmic governmentality, hybrid media circulation and emerging bottom-up social policies. It then shows how nominal values, real values at constant 2021 prices, year-by-year changes, moving-average smoothing, employment-scaled scenarios and transfer-balance indicators generate different representations of diaspora contribution, welfare substitution and national economic performance. Rather than assigning final authority to one dataset, the article demonstrates how calculation and presentation choices become communicative interventions. The conclusion emphasises methodological transparency and the need to connect remittance statistics to both political communication and community-level welfare practices. Full article

Phishing and malicious websites are a serious threat on the internet. Attackers use fake links to trick users and steal their private information. Detecting these links is difficult because attackers change their tricks often. Many old methods cannot detect new or hidden threats. Some recent models use deep learning (DL), but they are large, slow, and hard to use in real-time systems. In this paper, we present a lightweight and accurate model called MiniLM-CNNLSTM. It combines a small transformer model (MiniLM) with a hybrid DL network using Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) layers. The transformer learns the meaning of URLs. The CNN finds important patterns. The LSTM captures the order of characters. We also add handcrafted features that help the model detect tricky URLs. We test our method on two public datasets: the Phishing Site URLs dataset and the Malicious URLs dataset from Kaggle. We use 3-fold cross-validation and early stopping to ensure fair and stable results. The MiniLM-CNN-LSTM model outperformed previous benchmarks by achieving an average three-fold cross-validation accuracy of 98.98%, a precision of 98.63%, a recall of 98.29%, an F1-score of 98.46%, and a false positive rate of 0.68%. The proposed model has a higher accuracy, precision, recall, F1-score and a lower false positive rate, which enhances the accuracy by 1.88, precision by 3.77, recall by 4.17 and decreases the false positive rate by 61.58% compared with the strongest baseline (Distil BERT + CNN-LSTM), showing significant practical improvements. The results show that our approach is fast, small, and highly effective. It can detect phishing and malicious links with high accuracy. This makes it a good choice for real-time security systems like browsers, email filters, or firewalls. Full article

Radiomics enables quantitative medical image analysis by converting imaging data into structured, high-dimensional feature representations for predictive modeling. Despite methodological developments and encouraging retrospective results, radiomics continue to face persistent challenges related to feature instability, limited reproducibility, validation bias, and restricted clinical translation. Existing reviews largely focus on application-specific outcomes or isolated pipeline components, with limited analysis of how interdependent design choices across acquisition, preprocessing, feature engineering, modeling, and evaluation collectively affect robustness and generalizability. This survey provides an end-to-end analysis of radiomics pipelines, examining how methodological decisions at each stage influence feature stability, model reliability, and translational validity. This paper reviews radiomic feature extraction, selection, and dimensionality reduction strategies; classical machine and deep learning–based modeling approaches; and ensemble and hybrid frameworks, with emphasis on validation protocols, data leakage prevention, and statistical reliability. Clinical applications are discussed with a focus on evaluation rigor rather than reported performance metrics. The survey identifies open challenges in standardization, domain shift, and clinical deployment, and outlines future directions such as hybrid radiomics–artificial intelligence models, multimodal fusion, federated learning, and standardized benchmarking. Full article

Autonomous mobility is reshaping how travel time is perceived, experienced, and monetised. Most existing studies have examined the value of time (VOT), willingness to pay (WTP), comfort and safety perception, digital services, and user perception as isolated phenomena, with limited efforts to integrate these dimensions into unified analytical frameworks. This study aims to address the fragmented nature of existing research by developing an integrated understanding of user behaviour in autonomous mobility, linking VOT, WTP, psychological constructs, and service-related factors within a unified analytical perspective. A systematic review methodology following PRISMA 2020 guidelines was applied. A total of 81 peer-reviewed studies published between 2015 and 2026 were included and analysed, focusing on Private Autonomous Vehicles (PAVs) and Shared Autonomous Vehicles (SAVs). The results reveal three main trends. First, autonomous travel introduces greater flexibility in time use and enables productive or leisure activities during travel. Second, behavioural aspects of VOT and WTP are strongly influenced by psychological constructs such as trust, safety, and risk perception. Third, notable differences emerge between PAV and SAV contexts, particularly in terms of comfort, control, and safety perception. The literature predominantly employs stated preference surveys, discrete choice models, and hybrid models incorporating psychological factors. However, fragmentation persists in modelling behavioural aspects of time perception and shared mobility services. This study provides a structured synthesis of existing evidence and highlights key research gaps by integrating economic, psychological, and service-related dimensions. The findings emphasise the importance of context-specific and psychologically informed modelling approaches to better understand user acceptance and behavioural adaptation in autonomous mobility systems. Full article

Recognition of the American Sign Language (ASL) alphabet is of utmost importance in bridging the communication gap between the hearing-impaired and the hearing. However, robust classification remains difficult because some hand gestures are morphologically very similar. To address this problem, this study presents CT-Net, a hybrid deep learning architecture that integrates ConvNeXt-Tiny with a lightweight Transformer encoder. CT-Net combines convolutional feature extraction and self-attention mechanisms, which enable it to capture fine-grained local patterns and long-range spatial dependencies effectively. The proposed model was extensively compared with various architectures including traditional CNNs, Transformer-based models, hybrid machine-learning approaches and recent lightweight hybrid networks. The experimental results show that CT-Net achieved the best overall performance with a peak accuracy of 95.67% on the enhanced ASL dataset. Ablation studies demonstrate the effectiveness of our design choice. CT-Net achieves a strong trade-off between recognition accuracy and computational efficiency with an inference rate of 163.55 Frames Per Second (FPS). These findings highlight the potential of hybrid frameworks as a powerful tool for fine-grained gesture recognition tasks. Full article

As digital consumption increasingly unfolds in hybrid phygital environments, algorithmic systems play a growing role in shaping user choices, perceptions of fairness, and sustainability-related behaviour. Prior research has examined sustainable consumption, digital nudging, platform trust, and consumer behaviour in digital settings, but has rarely integrated perceived algorithmic fairness, digital resilience, and algorithmic responsibility perception within a single user-centered framework. Addressing this gap, this study develops and tests a multidimensional model of sustainable platform behavior (SPB). Using a triangulated design that combines bibliometric support analysis, PLS-SEM modelling, multi-group analysis, and cluster-based user segmentation, the study identifies three distinct user types and examines the relationships among the focal constructs. The results show that perceived fairness significantly predicts ARP (β = 0.493, p < 0.001), while both ARP (β = 0.427, p < 0.001) and digital resilience (β = 0.263, p < 0.001) independently contribute to SPB. The findings indicate that sustainable platform behavior is shaped not only by intention, but also by fairness perceptions, adaptive user capacity, and responsibility-based evaluations of platform systems. The study offers a user-centered framework with practical implications for designing more responsible, transparent, and sustainability-oriented digital platforms. Full article

Driven by the rapid growth of the low-altitude economy, drone logistics is emerging as a critical component of modern smart logistics systems. This study aims to examine how heterogeneous logistics service providers (LSPs) select among technological innovation, inter-firm cooperation, and hybrid strategies, as well as how these strategic choices affect ESG performance. We develop a two-stage duopoly Cournot game model that accounts for asymmetric logistics capabilities and consumers’ service-quality sensitivity, and compare the three strategic arrangements against a benchmark scenario without innovation or cooperation. Results show that a capability-driven Matthew effect already exists in the benchmark market. Technological innovation may further widen the performance gap between firms, yet it generates the highest social welfare by improving service quality and preserving market competition. Pure cooperation enhances coordination efficiency and environmental performance, but may reduce consumer surplus by weakening competition. The hybrid strategy generally delivers the highest system profit and robust environmental performance, while its advantages depend on market parameters and require sound benefit-sharing governance mechanisms. This study contributes to sustainable drone logistics research by integrating strategic interaction, firm heterogeneity and ESG outcomes into a unified framework, and provides targeted managerial and policy implications for innovation support, alliance governance and competition regulation. Full article

Existing soil and water conservation standards suffer from fragmentation, inconsistent cross-referencing, and limited machine interpretability, hindering efficient regulatory compliance and decision making. To address these challenges, we developed SwacGPT, an intelligent system that integrates domain-specific knowledge graph construction with large language models for enhanced standard interpretation and reasoning. Specifically, we constructed a domain-specific knowledge graph (SwacKG) using a hybrid approach that combines rule-based templates with a pre-trained BERT-based model. This graph systematically organizes conservation standards via multi-dimensional semantic relationships, with 87.8% entity extraction precision and 84.9% relation extraction precision, enabling precise data association across heterogeneous regulatory sources. SwacGPT leverages both the graph-structured knowledge from the SwacKG and original textual content to provide intelligent reasoning capabilities. For rigorous validation, a comprehensive evaluation dataset comprising both objective and subjective questions was designed. Experimental results show that SwacGPT achieves scoring rates of 78.67% on single-choice questions, 81.65% on multiple-choice questions, and 80.5% on subjective short-answer questions, ranking the best among the other five evaluated models. This demonstrates that the synergistic integration of domain-specific KGs with tailored LLMs creates an effective solution for intelligent environmental governance, providing critical decision support for land space optimization and cross-jurisdictional coordination in sustainable land management. Full article

Security monitoring in Industrial Internet of Things environments requires telemetry that spans Information Technology (IT) and Operational Technology (OT) network layers, and most public datasets capture only one such view. We describe a design pattern for hybrid Security Information and Event Management (SIEM) deployments in OT environments (rule-based detection plus edge-deployed machine learning anomaly detection writing into a shared index) and validate it on a Modbus/Jetson/Elastic instance. The pattern is platform-independent: any rule engine that exposes a query language and any edge device with adequate memory headroom can host an instance, and the paper documents the architectural choices that make this portability concrete. The validated instance comprises 27 rules in Kibana Query Language mapped to MITRE Adversarial Tactics, Techniques, and Common Knowledge, plus a CNN-BiLSTM autoencoder on a Jetson Orin Nano that reaches a true positive rate of 1.000 at the 98th-percentile validation threshold and 0.997 at the 99.5th-percentile threshold on a 9997-flow held-out attack partition. Runtime behaviour on the edge hardware is characterised under steady state and adversarial burst, including the queue-wait regime that dominates tail latency. A self-contained calibration step projects rule and model evidence onto a common scale for downstream fusion. Full article

The energy transition of district heating systems in Poland requires the simultaneous consideration of energy efficiency, operating costs, technical feasibility, and local environmental constraints. This study addresses an identified gap in the literature by combining real operational time series from a municipal district heating system with time-resolved market signals and site-specific resource constraints in a single OPEX-based operational screening framework. A case study is conducted for the city of Ustka using a configuration-based comparison of hybrid supply systems that include a gas-fired combined heat and power (CHP) unit, air-source and ground-source heat pumps, thermal energy storage, and a peak-load boiler. The optimisation model was implemented in MS Excel using the GRG Nonlinear algorithm (Solver) and was driven by the district heating operational data for 2021–2022 together with electricity and natural gas prices from the Polish Power Exchange day-ahead market (TGE RDN), evaluated under both hourly and daily settlement assumptions. The results indicate an optimal capacity split of 1.2 MWel/1.3 MWth for the CHP unit and 1.5 MWel/3.0 MWth for the heat pump system, supported by a required peak boiler capacity of 8.23 MWth. Within the adopted OPEX-based assessment, the lowest value of the unit heat generation indicator was obtained for the CHP-led configuration with combined ground-source and air-source heat pumps (38.45–38.55 PLN/GJ). A distinctive element of the study is the explicit verification of whether an operationally favourable configuration remains practically feasible when local resource constraints are considered. The site assessment indicates limited practical feasibility of the borehole heat exchanger at the analysed location in Ustka, showing that the lowest OPEX result should not be interpreted as a final investment recommendation. The study provides a replicable approach for the Polish district heating operators to screen hybrid transition pathways under real market conditions and to avoid technology choices that are favourable in dispatch models but constrained in practice. From a sustainability perspective, the proposed framework supports more energy-efficient, resilient, and locally feasible district heating transition planning in municipal heat systems. Full article

This study investigates the integration of categorical inverter alarm data into data-driven frameworks for photovoltaic (PV) system monitoring. While most existing approaches rely exclusively on continuous SCADA measurements, the potential of categorical operational data remains largely unexplored. In this work, categorical alarm signals are incorporated into power forecasting to enable anomaly detection. The proposed approach is evaluated on a large-scale real-world dataset comprising multiple PV plants and more than 100 inverters, representing over 1000 inverter-years of operation. The four most popular time series forecasting models, including Multi-Layer Perceptron, Long Short-Term Memory, Extreme Gradient Boosting, and Mamba, are used to estimate power output from continuous inputs, while categorical variables are integrated using one-hot encoding and entity embeddings. Anomaly detection is performed by analyzing residuals between predicted and measured power output. The results show that categorical alarm data contain relevant operational information and can be effectively incorporated into forecasting-based monitoring frameworks. However, their impact on predictive performance varies depending on the encoding strategy and model choice, highlighting important trade-offs between model complexity and feature representation. By providing a systematic evaluation of categorical data integration across a large, diverse dataset, this work addresses a gap in the literature and establishes a benchmark for future research on hybrid continuous–categorical approaches for PV inverter monitoring. Full article

This paper develops a deep supervised learning framework for sequential prediction of macro-financial time series derived from the U.S. Treasury yield curve. Using daily data on the 10-year Treasury constant maturity rate and the 3-month Treasury bill rate from the Federal Reserve Economic Data (FRED) system, we construct a sequence-based representation that captures level effects, yield spreads, short-term dynamics, and local volatility. We formulate the problem as a supervised learning task for predicting one-step-ahead changes in the 10-year Treasury yield. To model temporal dependence, we employ a hybrid convolutional–recurrent neural network that integrates convolutional layers for local pattern extraction and LSTM layers for capturing longer-range temporal structure. To improve training efficiency and robustness under nonstationary financial conditions, we investigate three experience replay strategies: uniform sampling, entropy–variance-based sampling that incorporates predictive uncertainty, and prediction-error-based sampling that prioritizes high-residual observations. Empirical results show that the choice of sampling strategy significantly affects both learning stability and out-of-sample predictive performance. Uniform sampling yields the most stable and competitive performance. Entropy-based sampling achieves strong late-stage predictive accuracy, indicating the benefit of uncertainty-aware data selection. In contrast, prediction-error-based sampling accelerates early learning but exhibits higher variance and reduced stability under structural breaks and regime shifts in macro-financial dynamics. Overall, the findings highlight a trade-off between stability, convergence behavior, and sensitivity to informative but potentially noisy observations in sequential financial prediction problems. The results suggest that simpler sampling strategies can be highly competitive in macro-financial environments characterized by nonstationarity and heteroskedasticity, while uncertainty-aware sampling provides a promising direction for improving predictive performance under uncertainty. Full article