Search Results (10,330)

Financial fraud is still one of the most important threats to the financial industry, causing enormous economic losses and mounting difficulties for conventional fraud detection systems. The systems tend to face challenges in dealing with the rising amount of transactional data, the problem of class imbalance, and the continually changing nature of fraudulent activity. In order to solve these problems, in this research a cloud hybrid framework for detecting fraud using Long Short-Term Memory (LSTM) networks, Autoencoders, and Cat Swarm Optimization (CSO) is suggested. The purpose of the suggested framework is to provide improved detection performance and flexibility on a benchmark financial dataset, with a design intended to support scalability in real-time applications. The framework uses the Credit Card Fraud Detection Dataset from Kaggle, which consists primarily of numerical features, including anonymized variables (V1–V28), along with time and amount. The LSTM networks learn the sequential relationships of transactions, while Autoencoders learn to detect anomalies in the data unsupervised. CSO is used to optimize key hyperparameters of the hybrid model, including the learning rate (0.0001–0.01), batch size (32–128), number of LSTM layers (1–3), number of hidden units per layer (16–128), dropout rate (0.1–0.5), and fusion weights (0–1 for each weight, with the sum constrained to 1) between the LSTM and Autoencoder outputs. In addition, CSO is applied for feature subset selection and threshold tuning to further enhance model performance. Preprocessing is performed on the data, including normalization and feature scaling prior to model training. The suggested framework has a 96.2% accuracy, 94.6% precision, 97.9% recall, 96.2% F1-score, and 0.97 AUC-ROC, showing improved performance compared to CNN-based and LSTM-CNN models under the evaluated conditions. However, since no multiple experiments were conducted to verify the robustness, the results should be interpreted as indicative rather than definitive. The framework exhibits competitive fraud detection performance on the evaluated benchmark dataset, particularly in handling class imbalance. In a simulated environment configured to mimic cloud-like conditions, the framework achieved inference latency between 15 and 30 ms, GPU utilization between 60% and 70%, and a data transfer volume of approximately 1.5 GB per day, suggesting its potential for deployment in cloud-based fraud detection systems. The framework indicates immense potential for cloud deployment, with a robust solution for preventing financial fraud. The proposed framework demonstrates the potential of integrating sequential modeling, anomaly detection, and metaheuristic optimization within a unified and cloud-oriented architecture, providing a more comprehensive approach compared to conventional hybrid models. Full article

This study investigates the structural misalignment between official wage benchmarks and actual market wages in the Republic of Korea to establish an independent, forestry-specific wage system essential for labor sustainability. Historically, the Republic of Korea forestry project costs have relied on construction industry benchmarks, leading to a “diverging hypothesis” where official rates fail to reflect the specialized risks and technical skills required in forest operations. To address this, a comprehensive wage survey was conducted in 2025 across 13 specialized forestry occupations. Utilizing a sampling frame of 7555 sites, 1044 units were selected via stratified sampling with square-root proportional allocation, ensuring a relative standard error (RSE) of 2.5%. The findings reveal that market wages consistently exceed construction benchmarks by 4.5% to 41.0%. The most significant disparities were observed in leadership and mechanized roles, reflecting substantial “risk–responsibility” and “skill premiums”. Furthermore, the study identifies a structural shift toward risk-transfer strategies, such as stumpage sales, in response to the Serious Accidents Punishment Act (SAPA). These results underscore the urgent need for a specialized wage framework to ensure safety and long-term resilience. Ultimately, such institutional refinement is a prerequisite for securing the high-quality human capital necessary for a sustainable circular bioeconomy. Full article

Social Network Analysis (SNA) has been widely applied to risk identification research. However, two key constraints, namely rating bias and the subjectivity of threshold selection, undermine the reliability and reproducibility of analytical results. To address this di-lemma, this study aims to construct a standardized and robust analytical framework for SNA-based risk identification. The core research objectives are as follows: elucidate the differential impact mechanism of threshold variation on the macro-topological structure and micro-level node ranking of risk networks, examine the cross-threshold robustness of core risk node rankings, and delimit the effective threshold range for stable risk identification. Accordingly, to fulfill the above objectives, this study proposes a multi-threshold robustness inspection method based on individual rating patterns, and conducts systematic empirical analysis with industrial projects in the post-support period of reservoir resettlement as research cases. The results indicate that threshold variation exerts marked systematic effects on the macro-topological structure of risk networks, whereas the relative rankings of core risk nodes remain robust. The effective threshold range for risk identification in such projects is α ∈ [0.1,0.3]. This study provides a repeatable quality control framework for SNA-based risk identification, with favorable cross-domain transferability. Full article

Non-invasive temperature estimation during online operation is a critical challenge in enclosed micro-reaction systems, particularly when the thermal mass of the working fluid varies dynamically or is uncertain. Conventional model-based approaches typically rely on fixed thermal parameters, leading to significant estimation errors when the actual reagent volume deviates from nominal conditions. To address this limitation, this study proposes a volume-adaptive temperature estimation framework applied to an ultrafast quantitative polymerase chain reaction (qPCR) system. By modeling the heat-transfer pathways via a simplified resistance–capacitance (RC) network, a nonlinear least squares (NLS) algorithm within an output-error (OE) framework is employed to identify key thermal parameters online. The framework separates the estimation into an offline calibration stage—where a thermocouple-equipped chip provides ground-truth data—and an online deployment stage that relies solely on non-invasive external measurements. This approach allows the system to explicitly compensate for volume-induced variations in thermal inertia. Validation experiments on an ultrafast qPCR platform with reagent volumes ranging from 100 to 250 $\mathsf{\mu }\mathrm{L}$ and heating rates exceeding 20 °C/s demonstrate that the method achieves robust performance, maintaining a mean absolute error (MAE) of reagent temperature at 0.24 ℃ and restricting the average volume estimation error to within 1.37 $\mathsf{\mu }\mathrm{L}$. DNA gel electrophoresis results further confirm the biological reliability of the temperature prediction strategy by verifying amplification specificity. This work provides a generalised solution for precise thermal management in micro-systems subject to variable thermal loads. Full article

Uncrewed aerial systems (UASs) have diverse applications in natural environments, yet their deployment can inadvertently disturb wildlife. This PRISMA-guided systematic review synthesised 39 studies (2015–2025) encompassing birds, mammals, and marine taxa to identify UAS operational drivers of disturbance. We applied a Factors–Responses–Consequences (F–R–C) framework linking UAS operational characteristics, observed wildlife responses, and ecological consequences. Three patterns emerged: (i) Factors: Contextual and operational conditions such as flight altitude, noise, and approach direction interact with species-specific sensitivities to shape disturbance potential. (ii) Responses: Physiological measures (e.g., elevated heart rates) often reveal hidden stress not evident from behaviour alone. (iii) Consequences: Short-term effects may accumulate into long-term impacts on health, reproduction, and habitat use. These findings highlight the need for species- and context-specific flight envelopes rather than solely uniform altitude limits. By structuring existing evidence within the F–R–C framework, this synthesis provides a transferable foundation for UAS mission planning, drone development, operational decision-making, ethical practice, and environmental impact assessment in conservation and wildlife-management contexts. As all screening and data extraction were conducted by a single reviewer, the findings should be interpreted with appropriate caution pending independent validation. Full article

Cadmium is an element that is unnecessary for the functioning of plant and animal organisms, and its widespread presence in the environment poses a serious threat to human and animal health. Therefore, effective methods are being sought to remediate soils contaminated with this element, including through the enrichment of degraded soils with organic matter. To this end, the effectiveness of selected organic sorbents, including starch, fermented bark, compost and humic acids, in mitigating the transfer of cadmium and other heavy metals from soil to plants was assessed. Model studies compared the effects of 15 and 30 mg of cadmium (Cd) per kg of soil with an uncontaminated control sample. The sorbents were applied on a carbon basis at a rate of 3 g C per kg of soil. The test plant was Zea mays. Cadmium was found to significantly impair plant growth, causing reductions of 21%, 85%, and 77% in leaf greenness, aboveground biomass and root biomass, respectively. Excess cadmium increased the translocation of lead, chromium, copper, nickel, zinc, iron, and manganese from the roots to the aboveground parts of the plant, while simultaneously limiting their uptake. All of the organic sorbents tested reduced the negative impact of cadmium on leaf greenness, except starch. Compost and HumiAgra significantly improved the condition of Zea mays plants weakened by cadmium exposure. Cadmium contamination increased soil acidification. pH was positively correlated with maize yield and the SPAD leaf greenness index and negatively correlated with the cadmium translocation index and cadmium content in the aboveground parts of maize. Compost and humic acids are among the most effective and practically feasible approaches for reducing cadmium bioavailability in soil and its accumulation in Zea mays, and are therefore recommended for the remediation of cadmium-contaminated soils. Full article

The produced water from the No. 1 Oil Production Plant of Xinjiang Oilfield is rich in divalent ions, including Ca²⁺, Mg²⁺, and SO₄²⁻, leading to extremely high scaling tendency that fails to meet the reinjection standard. Therefore, highly efficient water softening technology is urgently required for such wastewater treatment. In this study, a novel negatively charged nanofiltration (NF) membrane was fabricated via interfacial polymerization using 2-carboxypiperazine and trimesoyl chloride as monomers. The membrane was systematically characterized by scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS), and Fourier-transform infrared spectroscopy (FTIR), and its rejection performance was investigated under various conditions. Results show that the maximum rejection rates of the NF membrane reached 99% for SO₄²⁻, 81% for Ca²⁺, and 94% for Mg²⁺, respectively. With increasing ion concentration, the removal efficiencies of Ca²⁺ and Mg²⁺ decreased, while that of SO₄²⁻ increased slightly. Higher operating pressure significantly enhanced both ion removal and membrane flux, which was mainly attributed to the synergistic effects of Donnan electrostatic exclusion, membrane surface adsorption, and mass transfer resistance. When applied to treat real produced water from the No. 1 Oil Production Plant, the membrane achieved 100% removal of SO₄²⁻, and 91% and 95% removal of Ca²⁺ and Mg²⁺, respectively. The scaling tendency of the treated effluent was completely eliminated. This work provides theoretical and technical support for the engineering application of nanofiltration technology in oilfield wastewater treatment. Full article

Slowloris is a well-known application-layer Denial of Service (DoS) attack that is challenging to detect due to its low-rate nature, allowing it to blend with legitimate traffic and remain unnoticed. Our hypothesis is that deliberate prolongation of the pre-authentication stage in stateful protocols induces unnecessary CPU utilization. In this study, we repurpose Slowloris as an energy-oriented (E-DoS) attack that exploits pre-authentication statefulness of the most prevalent remote access protocols, the Secure Shell Protocol (SSH) and File Transfer Protocol (FTP). We employ a Raspberry Pi-based experimental setup with different software implementations of the mentioned protocols to validate our hypothesis. Our experiments confirm the susceptibility of SSH and FTP to Slowloris E-DoS attacks, and we quantify the consequential impact on power consumption. We find that the Slowloris E-DoS attack exhibits an asymmetrical nature, causing a disproportionate computational demand on victim systems compared to the resources invested by the attacker. The results of this study indicate that battery-powered single-board computers (SBCs) are critically affected by these attacks due to their limited power availability. This research demonstrates the importance of understanding and mitigating Slowloris E-DoS vulnerabilities in the SSH and FTP protocols, offering valuable insights for enhancing security measures. Our findings show that millions of SBCs worldwide may be at risk and highlight a deeper structural weakness: the stateful design of widely deployed protocols can turn service availability into an energy liability. This systemic risk extends beyond SSH and FTP, with implications for IoT devices and backends that depend on stateful communication protocols. Full article

Air pollution remains a critical environmental and public health threat, particularly in highly populated urban areas such as the Lisbon Metropolitan Area (LMA). This study provides a refined and detailed assessment of the spatial distribution of air pollution and associated attributable mortality across the LMA. High-resolution (1 km²) annual mean concentrations of key pollutants (PM_2.5, PM₁₀ and NO₂) for 2022 and 2023 were estimated by integrating outputs from the URBAIR dispersion model with ground-based monitoring observations using advanced geostatistical data-fusion techniques. Air pollutant concentrations were combined with gridded population data and age-stratified baseline mortality rates within a Geographic Information System framework to quantify spatial variations in health impacts. Using the World Health Organization AirQ+ framework and established concentration–response functions, we estimated a total of 3195 air-pollution-attributable deaths across the Lisbon Metropolitan Area (LMA) in 2022, increasing to 4010 deaths in 2023. Fine particulate matter (PM_2.5) was identified as the dominant contributor, accounting for more than 40% of the total health burden. At a high spatial resolution (1 km² grid), estimated mortality exhibited substantial variability, ranging from 0 to 29 deaths per cell in 2022 and from 0 to 36 deaths per cell in 2023. These results highlight the importance of fine-scale spatial analysis, revealing intra-urban disparities that are not captured by aggregated estimates of total attributable mortality. The proposed methodological framework, integrating dispersion modelling, data fusion, and spatially explicit health impact assessment at fine spatial scales, provides a robust and transferable approach to support evidence-based air quality management and urban health policy development in European metropolitan contexts. This integrated approach enhances comparability, improves exposure assessment accuracy, and strengthens the scientific basis for designing targeted mitigation strategies that could prevent hundreds of premature deaths annually while addressing documented spatial inequalities in pollution exposure. Full article

To reveal how the characteristic flavor of jasmine tea is generated, this study analyzed the coordinated changes in sensory properties, chemical components, and aroma migration behavior during scenting. Sensory evaluation, biochemical assays, and headspace solid-phase microextraction–gas chromatography–mass spectrometry (HS-SPME-GC-MS) integrated with orthogonal partial least squares discriminant analysis (OPLS-DA) and relative odor activity value (rOAV) filtering were applied to tea samples before and after scenting. After scenting, aroma and taste scores increased significantly, and liquor color shifted from tender green to pale yellow. Amino acids and soluble sugars increased, while astringent substances such as tea polyphenols and catechins decreased. Key floral compounds, including cis-3-hexenyl benzoate and methyl anthranilate, were transferred from jasmine flowers to the tea base and enriched, likely contributing to the typical aroma profile. The retention rate of aroma in spent flowers was positively correlated with hydrophobicity (logP, r > 0.46, p < 0.01) and negatively with polarity (TPSA, r > −0.42, p < 0.05), suggesting regulation by hydrophobic partitioning. In contrast, aroma transfer to the tea base showed no simple correlation with any single physicochemical parameter, suggesting multi-factor regulation. This study provides insights into the scenting process and offers a theoretical reference for quality control in jasmine tea production. Full article

The nominal maximum aggregate size (NMAS) plays a critical role in determining the mechanical performance of cement-stabilized macadam (CSM), yet its meso-mechanical influence mechanism remains insufficiently understood. In this study, three skeleton-dense CSM mixtures with different NMAS values were designed, and a combined experimental–numerical approach was adopted to investigate the macro- and meso-scale mechanical behavior. Uniaxial compression tests and aggregate crushing value tests were conducted to evaluate strength development and load-transfer characteristics, while a three-dimensional discrete element method (DEM) model incorporating realistic aggregate morphology was established to analyze the evolution of contact forces and crack propagation. The results show that increasing NMAS significantly improves the mechanical performance of CSM. Compared with CSM-30, the 7-day compressive strength of CSM-40 and CSM-50 increased by approximately 10.3% and 37.3%, respectively. The stress–strain response indicates that mixtures with larger NMAS exhibit higher stiffness and a higher strain. At the meso-scale, a larger NMAS promotes the formation of a more efficient force-chain network dominated by coarse aggregates. Strong contacts were predominantly carried by aggregates larger than 9.5 mm, and in CSM-50, the proportion of strong contacts in the 37.5–53 mm fraction exceeded 90%, indicating that the largest particles likely form the primary load-bearing skeleton. In addition, increasing NMAS delayed crack initiation, reduced crack propagation rate, and decreased the total number of cracks at failure. These findings demonstrate that macroscopic strength improvement is closely associated with meso-scale optimization of the aggregate skeleton and enhanced load-transfer efficiency. This study provides a mechanistic basis for NMAS selection and gradation optimization in semi-rigid base materials. Full article

Online asynchronous learning offers considerable flexibility but frequently faces challenges in sustaining engagement, interactivity, and inclusivity across diverse learner populations. This study introduces the OPTIMAL framework—an Online, Pedagogy- and Technology-Integrated, Microcurricula Approach for interactive and inclusive Learning—synthesising universal design for learning, active learning, and constructive alignment with technology integration frameworks (TPACK and PICRAT), operationalised through a microcurricula-as-a-service architecture. A three-year longitudinal case study (2022/23 to 2024/25) examined the application of the framework to a data literacy and analytics module serving over 5000 students across more than 15 programs and five faculties at Dublin City University. The module design constructively aligned learning outcomes, content, and technology at three levels to support multiple learning pathways, formative assessment, and transdisciplinary engagement, deliberately fostering transformative uses of technology in a fully asynchronous environment. Mixed-methods evaluation—combining learning analytics, surveys (n = 1743), and qualitative feedback—demonstrated sustained positive outcomes across all three years, including 95–99% completion rates, consistently high satisfaction, and longitudinal gains in engagement and pass rates. These findings demonstrate how the deliberate integration of pedagogical theory, technological frameworks, and modular curriculum architecture can deliver scalable, inclusive, and high-engagement online education, offering both a transferable, evidence-based model for educators and curriculum designers and longitudinal empirical validation for researchers. Full article

Sensitive detection of Escherichia coli O157:H7 (E. coli O157:H7) in food matrices remains an important analytical challenge. Here, a colorimetric biosensor was constructed based on a bimetal oxide nanozyme composed of Ce-Fe oxide. This biosensor achieved sensitive detection of E. coli O157:H7. The Ce-Fe oxide synthesized on the basis of deep eutectic solvents (DESs) had the advantages of low solvent consumption and short preparation time. By regulating the two key factors of metal valence and oxygen vacancy content, the peroxidase (POD) activity of the nanozyme was significantly improved. Compared with the single-metal oxide nanozyme Fe oxide, the addition of Ce increased the Fe²⁺/Fe³⁺ ratio from 0.37 to 0.49, implying a possible enhancement of electron transfer between Fe²⁺ and Fe³⁺. The detection limits (LODs) of the biosensor based on Fe oxide and that based on Ce-Fe oxide were 10² CFU/mL and 10¹ CFU/mL, respectively, comparable to existing validated methods. Moreover, these two biosensors achieved satisfactory recovery rates (91–104%) and RSDs (1.2–8.8%) in the spiked lake water, juice, and lettuce samples of E. coli O157:H7, indicating their high potential for application in spiked sample detection. In summary, the method proposed in this study for improving the POD activity of nanozymes through electron transfer in DES solutions is beneficial to the development of metal oxide nanozymes. Full article

Achieving high selectivity in the hydrogenation of CO₂ to light olefins remains challenging because of the complex reaction network and the difficulty of regulating key intermediates. Motivated by green-hydrogen-enabled power-to-chemicals pathways, we combine density functional theory (DFT) with first-principles microkinetic simulation (FPMS) to construct a quantitatively predictive reaction-energy landscape and elucidate structure–selectivity relationships. A comprehensive reaction network is established through energy-surface fitting, and steady-state rate constants are solved to capture the microkinetic competition between elementary steps. By introducing electronic density-of-states (DOS) modulation as a design variable, we directly correlate surface structural parameters with rate-controlling steps, thereby enabling targeted regulation of C–C coupling and hydrogen transfer processes. The calculated barrier for CO₂ adsorption to COOH* is 1.35 eV, while the transition state barrier for C–C coupling is 1.50 eV, corresponding to a reaction rate of 9.7 × 10³ s⁻¹; the olefin desorption rate reaches 1.7 × 10⁷ s⁻¹. Crucially, shifting the d-band center from −2.35 eV to −1.60 eV increases the C₂–C₄ olefin selectivity from 42.6% to 68.3%, establishing an actionable electronic structure lever for catalyst optimization. These results reveal the intrinsic mechanism by which surface electronic and geometric regulation governs intermediate stabilization and rate control, providing a verifiable, mechanism-based design principle for efficient CO₂-to-olefin catalysts aligned with green hydrogen deployment. Full article

Lipid nanoparticles (LNPs) have emerged as a groundbreaking delivery platform, revolutionizing the development of nucleic acid-based medicines for gene delivery and gene therapy. This review provides an insightful industrial perspective on the production process of LNPs, focusing on cutting-edge manufacturing equipment, downstream processing and the crucial transition from laboratory to large scale. While LNP production in the discovery phase relies on a small scale (µL to mL) for screening various LNP formulation candidates, transferring to preclinical (up to hundreds of mL) and clinical/commercial scales (up to liters) requires a robust and reproducible manufacturing process. Thus, mixing technologies throughout these scales must be carefully selected and require precision, scalability and high reproducibility to meet the target quality of the LNP drug product. Key mixing technologies in mRNA-LNP production primarily include microfluidic systems and impinging jet mixers (IJMs). In this review, we discuss key critical process parameters (CPPs) in LNP preparation, including flow rate ratio (FRR) or total flow rate (TFR), in relation to associated critical quality attributes (CQAs) across multiple manufacturing scales. We further assess the impact of downstream processing, specifically tangential flow filtration (TFF), on the formulation’s CQAs. In particular, the review highlights the importance of maintaining CQAs along each step of the process and emphasizes the role of robust analytical methods in ensuring product quality and safety. Additionally, we touch on current challenges associated with these advanced delivery vehicles, such as their long-term stability, and introduce the readership to innovative stabilization strategies aimed to extent LNP shelf-life. Full article