Search Results (895)

Emerging-market equity exchanges require regime forecasting systems that are continuous in time, robust to heavy-tailed distributions, and optimised against false alarms. No existing method addresses all three simultaneously, and no prior study has reported a crisis false-alarm rate on JSE equities. We propose S-NODE-ANF-RRC: a stochastic neural ODE within an Adaptive Neuro-Fuzzy Risk-Regime Clustering architecture, integrated by a Milstein scheme with Lyapunov-regularised dual-loss training. The system is evaluated as a one-step-ahead probabilistic forecaster ($h=1$ trading day) on 2696 daily observations across 17 JSE securities (March 2015–March 2026). Gaussian mixture clustering on raw features (kurtosis 54.8) inflates ARI by 1.3×; log-transformation corrects this artefact. Two operational profiles emerge: the N-ODE-ANF-RRC achieves the lowest cost (10,350 bp, 65.1% below GMM) and longest lead time (0.71 days); the S-NODE-ANF-RRC achieves the lowest false alarm rate among probabilistic architectures (FAR = 0.051), with a 42.0% cost reduction versus GMM (McNemar $p=0.027$, power $1-\beta =0.73$; bootstrap CI [5250, 19,600] bp excludes zero). Ablation confirms drift, diffusion, and dual-loss as the minimum viable daily-frequency configuration. Full article

Introduction: Populations of freshwater fish species of the Iberian Peninsula have been declining since the mid-20th century, and several types of actions (from in situ to ex situ actions) have been applied over the past decades. However, limited knowledge about their reproductive biology makes it necessary to investigate different aspects of the reproductive cycle for improving breeding programs. Objective: The main objective of this work was to advance knowledge in the sperm biology of two endemic fish from Portugal rivers, trying to check whether breeding in captivity is a factor able to modulate sperm subpopulations. Methodology: Populations of different endangered leuciscid species (Iberochondrostoma lusitanicum, IL, and Achondrostoma occidentale, AO) were sampled during the spring of 2022 both in captive populations kept at Aquário Vasco da Gama (AVG) and in wild populations (WILD) from different Portuguese rivers. Sperm samples were collected, and sperm motion parameters were assessed by using a CASA system (VSL, VAP, STR, LIN, WOB, ALH and BCF). Results: The application of a two-step cluster analysis yielded four sperm subpopulations (SP1, SP2, SP3 and SP4) in both species. SP1 comprised fast, linear spermatozoa (high VCL, LIN, STR). SP2 contained slow linear cells (low VCL, high LIN, STR). SP3 included fast nonlinear spermatozoa (high VCL, low LIN, STR). SP4 represented slow nonlinear cells, with low values for all three kinematic parameters. Regarding the origin of fish (wild and captive), and for both species, WILD leuciscids showed higher values of linear and fast sperm subpopulation (SP1) than captive fish (AVG), which showed a higher percentage of non-linear subpopulations (SP3 and SP4). Conclusions: In this context, and given that fast and linear spermatozoa (SP1) have traditionally been correlated with high fertilization success in many fish species, these results may indicate that breeding in captivity over a long period of time may affect gamete quality, making it necessary to renew the broodstock from time to time to avoid reproductive problems (i.e., loss of sperm quality and cases of inbreeding). Full article

Intercultural education in Mapuche contexts is shaped by persistent tensions between dominant school knowledge and Indigenous educational practices. However, there is limited comparative empirical evidence on how these tensions are distributed across educational actors. This study aimed to compare socio-educational and cultural ambivalence between students and teachers across multiple dimensions. A cross-sectional quantitative design was conducted with 546 participants (284 students and 262 teachers) from primary and secondary schools in southern Chile. Ambivalence was assessed using the Socio-Educational and Cultural Ambivalence Scale (EASC). A two-step cluster analysis identified ambivalence profiles, followed by a 2 × 2 factorial MANOVA (role × ethnicity). Results revealed three distinct ambivalence profiles (low, medium, high), with significant differences across all dimensions (p < 0.001). Multivariate analyses showed significant effects of role (Pillai’s trace = 0.230, F (6, 537) = 26.67, p < 0.001, η²p = 0.230) and ethnicity (Pillai’s trace = 0.108, F (6, 537) = 10.86, p < 0.001, η²p = 0.108), with no significant multivariate interaction (p = 0.104). Teachers reported higher levels of ambivalence than students in five of six dimensions, while Mapuche participants scored higher than non-Mapuche participants across most dimensions. These findings indicate that ambivalence is a structural condition of the educational system, unevenly distributed according to actors’ positions and intensified in roles involving pedagogical mediation. Implications point to the need for structural transformations in intercultural education, particularly in teacher education. Full article

Iron metabolism has long been linked to metabolic syndrome (MetS), but it is still unclear at which step—iron sensing, hepcidin regulation, export, transport, or storage—genetic variation matters the most. There are almost no studies on iron metabolism genes in Kazakhs in particular. Using whole-exome sequencing (WES) data from 96 Kazakh adults (52 with MetS), we examined 18 SNPs across six iron metabolism genes—HFE, SLC40A1, TMPRSS6, FTL, TFR2, and TF. Associations with iron biomarkers and MS components were tested by linear regression adjusted for age, sex, and BMI, with FDR correction, haplotype analysis, and bootstrap mediation analysis. Significant effects clustered at two distinct steps of iron metabolism: hepcidin regulation (TMPRSS6) and iron transport (TF). The T allele of TF rs12769 raised serum transferrin (β = +0.32 g/L; p_FDR = 0.002) while lowering both TSAT (β = −4.25%) and ferritin (β = −0.36 log-units); haplotype analysis confirmed rs12769 as the driver. The TMPRSS6 C–G–C haplotype was associated with lower fasting glucose (β = −1.19 mmol/L; p = 0.023), and TF rs12769 emerged as a robust FDR-significant determinant of serum transferrin (p_FDR = 0.002). Bootstrap mediation analysis (5000 iterations) showed that the TMPRSS6 effect on glucose is not mediated by ferritin, serum iron, transferrin, TSAT, or sTfR (all ACME p > 0.20), while Total and Direct Effects remained robust (p ≤ 0.054). In Kazakhs, iron-metabolism genes appear to influence fasting glucose through direct mechanisms not captured by the standard iron biomarker panel; alternative pathways involving hepatic enzymes, hepcidin, or inflammation warrant investigation in larger cohorts. Full article

Microsatellites operate in highly dynamic thermal environments due to severe physical constraints, making temperature telemetry a critical onboard health indicator. Conventional threshold-based monitoring fails to distinguish normal operational mode transitions from genuine faults, causing excessive false alarms. To address this, we propose a two-stage framework integrating unsupervised thermal mode discovery with mode-specific deep learning prediction. Raw temperature telemetry is downsampled and segmented into orbital cycles. Unsupervised clustering identifies two nominal thermal regimes and four canonical fault-type libraries (step, spike, drift, and noise), each corresponding to distinct in-orbit failure mechanisms. For each nominal mode, a Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) is trained on 7-day historical windows to forecast 3-day temperature evolution. Post-downlink, incoming cycle mode is inferred via nearest-neighbor DTW classification; anomalies are flagged when prediction residuals exceed mode-adaptive thresholds. Validation on Macau Science Satellite-1B (MSS-1B, COSPAR 2023-069-B, NORAD 56732) in-orbit telemetry from a 41° inclination low-Earth orbit—where solar illumination dominates external thermal loading and internal heat from the data-communication module and scientific payload constitutes the primary internal thermal source—shows the method reduces anomaly flags by 96.6% and improves prediction mean absolute error by 51.3% compared to a non-classified global baseline under nominal operating conditions, correctly detecting a known operational transient while suppressing spurious alarms. A synthetic fault injection experiment with four anomaly types and five baseline methods further confirms the framework’s detection capability, achieving an overall F1 score of 0.725 vs. 0.258 for the global baseline—a 2.8× improvement driven primarily by a 4× precision gain. Sensitivity analysis reveals that the two-stage advantage is most pronounced for low-magnitude and short-duration faults, where mode-specific context is essential. This work advances microsatellite autonomous health management by providing reliable anomaly detection with quantified fault detection performance. Full article

For the identification problem of Hammerstein models, this paper proposes a decoupled identification algorithm that combines the correlation analysis with the optimal gradient-based iterative algorithm. The proposed method firstly decouples the linear and the nonlinear subsystems of the Hammerstein model with the help of the correlation analysis. Then, the least squares algorithm is used to estimate the linear parameters. For the nonlinear subsystem, which is represented by a fuzzy neural network, a clustering algorithm is employed to initialize the parameters of the membership functions, so that the fuzzy rules can better reflect the distribution characteristics of the observation data. In the weight identification stage, the optimal gradient theory is introduced into the gradient-based iterative algorithm. Based on the solved optimal step size, the method achieves adaptive parameter updates, which significantly improves convergence speed while maintaining algorithm stability. Finally, the proposed algorithm is applied to two simulations, and the results demonstrate its effectiveness for both linear and nonlinear parameters in Hammerstein models. Full article

Ultra-short-term electricity load forecasting is crucial for real-time power system operation, but its accuracy is limited by the nonstationary and multiscale characteristics of load data. To address this issue, this study proposes a multi-scale decomposition–clustering and heterogeneous gated fusion framework. The original load sequence is decomposed by ICEEMDAN and then grouped into high-, mid-, and low-frequency components using K-means clustering. MS-gTCN is used to capture high-frequency fluctuations, adaptive DLinear is used to model low-frequency trends, and a gated fusion mechanism is designed for mid-frequency components. A lightweight error correction network is further introduced to reduce residual prediction errors. Experiments on two real-world datasets show that the proposed method achieves the best performance across 1-, 4-, 8-, and 12-step horizons. For the 12-step task, it reduces MAE by 29.3% on Dataset A and 26.2% on Dataset B compared with the second-best baselines. Compared with ICEEMDAN-LSTM on Dataset A, it reduces MAE by 17.7% and improves $R^2$ from 0.9127 to 0.9418. Ablation, sensitivity, significance, and complexity analyses further verify the effectiveness, robustness, and real-time feasibility of the proposed framework. Full article

Pre-classification of lakes into optical water types (OWTs) is considered a useful step in analyzing satellite-based reflectance data. We used a dataset of 109 reflectance hyperspectra from Minnesota and Wisconsin lakes and rivers to evaluate the usefulness of pre-classification to improve the retrieval of water quality information from satellite data. Three OWT classes were derived from the dataset by K-means clustering using three integrative metrics of reflectance spectral shape and magnitude as clustering variables. Values of the three metrics can be determined from satellite reflectance data as well as hyperspectral data. The OWT classes had distinct water quality characteristics in terms of Secchi depth, chlorophyll-a, and colored dissolved organic matter (CDOM). Algorithms used to retrieve values of the variables from simulated Sentinel-2 band reflectance data usually yielded more accurate predictions when computed separately for each class than when computed for the entire dataset, although exceptions were found for some fitting metrics and models and results for chlorophyll-a were not definitive. The three water quality variables were related in distinct ways to the integrative shape metric of reflectance spectra, apparent visible wavelength (AVW), supporting its use to develop OWTs to organize waterbodies into water quality classes. AVW was correlated (r = 0.933) with the integrative metric, normalized difference index at green and red wavelengths (NDI). Based on that result, we found that OWTs developed using just two variables, AVW and a metric of spectral magnitude, were nearly the same as classifications using all three integrative metrics. Full article

We introduce the symmetric co-skewness moment (SCKM)—a third-order informational dissimilarity metric that consistently outperforms state-of-the-art client-selection heuristics in federated learning (FL) under heterogeneous data. Unlike similarity-driven schemes, SCKM minimizes redundancy by favoring clients with complementary gradients, delivering faster and more stable convergence even at high heterogeneity levels. Operating on highly compressed 0.5% gradient summaries, our framework provides two operating modes for different deployment scales: (i) SCKM-Select directly ranks and schedules a small candidate pool, whereas (ii) SCKM-Cluster adds a fast, elbow-guided clustering step to scalably choose from thousands of users. We evaluate both variants on a VGG-16 model across multiple non-IID partition schemes and initializations, observing consistent gains over leading cosine-similarity, loss-sketch, and max-diversity baselines—without increasing the communication budget. Full article

Vapor Pressure Deficit (VPD) is a critical determinant of atmospheric evaporative demand and plant water stress in tropical agricultural systems. This study applied a Gaussian Mixture Model (GMM) and K-Means clustering to 36,528 hourly meteorological observations collected from Eastern Thailand between August 2021 and September 2025, with the objective of identifying distinct atmospheric moisture regimes relevant to precision irrigation management in durian cultivation. Two input configurations were evaluated: a multivariate feature space comprising air temperature, relative humidity, wind speed, solar radiation, and VPD; and a univariate input consisting of VPD alone. Model selection for GMM was guided by the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC), while K-Means performance was assessed using the Elbow method, Silhouette Coefficient, Calinski–Harabasz Index, and Davies–Bouldin Index. For the multivariate input, GMM identified K = 7 as the optimal number of clusters, supported by the largest single-step reduction in both AIC and BIC at this transition point. For the univariate VPD input, K = 5 was selected as the most parsimonious and agriculturally interpretable solution. The seven clusters derived from the multivariate GMM were organized into four atmospheric moisture regimes, such as very low, moderate, high, and very high evaporative demand, capturing the full spectrum of diurnal and seasonal VPD variability characteristic of Eastern Thailand. The results demonstrate that GMM-based probabilistic clustering applied to multivariate meteorological inputs provides a more comprehensive characterization of atmospheric moisture dynamics than univariate or geometric clustering approaches, offering a practical framework for tiered irrigation scheduling and drought stress early warning systems in tropical fruit cultivation. Full article

Nature-based tourism is expanding rapidly, placing new pressures on fragile ecosystems and governance structures that were not designed for the intensity and diversity of today’s visitors. Despite this trend, protected areas face unique management constraints and rapid socio-environmental changes. While motivational segmentation of tourists can provide valuable information to policymakers, this subject remains understudied/under-researched. This study addresses the gap by examining the motivations, behaviours, and attitudes of visitors to Romania’s national and natural parks, using a structured survey (n = 509) and a two-step approach combining dimensionality reduction with visitor segmentation. Principal component analysis (PCA) reveals distinct motivational dimensions related to visitors’ desire for immersion in nature, wildlife observation and learning, active recreation, and social–cultural engagement. Based on these dimensions, three visitor segments emerge through cluster analysis, with significantly different patterns of landscape use, expectations of recreational services, and perceptions of interpretation media. This research provides practical insights for targeted communication, zoning, and adaptive governance and proposes integrating visitor typologies with park management to support sustainable rural development. The findings highlight how a nuanced understanding of tourist segments can inform more effective policy measures that balance recreational demand with the long-term protection of natural and cultural resources, offering practical value for the sustainable development of protected areas, local communities, and other stakeholders. Full article

Efficient textile recycling depends on accurate identification of fibre types and compositions to support high-value material recovery and automated sorting. Existing commercial systems based on near-infrared (NIR) spectroscopy offer robust performance, but their model architectures and development methods are proprietary, and they often struggle to detect materials when carbon-black (graphite-based) dyes suppress the spectral signatures. This paper presents a hyperspectral imaging approach for textile fibre identification, combined with an artificial intelligence model capable of detecting cotton, polyester, elastane, and regions affected by carbon-black dye. Sixty-five textile samples were laboratory-verified to determine constituent materials and compositions, with 52 used in model development and testing. A semi-automatic algorithm detected textile boundaries and sampled 100 spectral patches per image. For materials exhibiting two distinct spectral signatures, typically due to carbon-black dye regions, 100 samples were collected for each signature, producing a database of 6500 spectra. A convolutional neural network model was trained using these signatures to predict fibre composition and identify any regions with carbon-black dye. The system achieved mean absolute errors below 2.1% for cotton, polyester, and elastane. A spatial clustering step groups pixels with similar spectra prior to detection, enabling region-wise material identification and allowing the model to classify clusters likely affected by carbon-black dye. This approach demonstrates high precision in fibre identification and reliable detection of carbon-black regions, highlighting its suitability for real-world textile analysis workflows. Full article

The catalytic isomerization of glucose to fructose is a pivotal step in the valorisation of lignocellulosic biomass. In this study, polyaluminium chloride (PAC), a low-cost, industrially available material characterised by polynuclear aluminium–oxo species, is investigated for the first time as an alternative catalyst for this transformation. The catalytic performance of PAC was systematically investigated by varying the temperature (70–130 °C), solvent (pure water, H₂O:MeOH 1:1 or 4:1) and reaction time (30–120 min). A fructose yield of up to 55% with a selectivity of 85% was obtained by using PAC at 120 °C in H₂O:MeOH 1:1 for 120 min, confirming its effectiveness in promoting glucose isomerization to fructose. Mechanistic insights and quantitative monitoring were achieved using benchtop NMR spectroscopy. ²⁷Al-NMR of PAC in aqueous solution exhibits two main signals at 1.12 ppm (attributed to the hexacoordinated Al complex) and at 63.3 ppm (associated with a tetrahedral Al centre typical of the Al₁₃ Keggin-type cluster). With the increase in temperature, as well as by changing the reaction media from pure aqueous to a mixed aquo-alcoholic system, new Al species were generated that are more reactive than the starting AlCl₃∙6H₂O and PAC. Overall, this work demonstrates that PAC represents a viable, scalable, and more sustainable alternative to conventional aluminium-based catalysts, offering a promising route toward more efficient biomass conversion processes. Full article

Narrative extraction algorithms construct storylines by finding coherent paths through document collections. The Narrative Trails algorithm frames this as maximum-capacity path optimization, where path quality depends on a coherence function measuring document relationships. We introduce quantum kernels as coherence functions for narrative extraction—to the best of our knowledge, the first systematic characterisation of quantum kernel methods for storyline extraction—and compare them against classical baselines on two corpora using a multi-seed protocol. The sweep covers 93 method evaluations (54 quantum kernels across three encoder families—R_Y+CNOT-ring, IQP/ZZ-feature-map, and a projected quantum kernel—and 39 classical kernels—cosine, RBF, and the cluster-aware Narrative Trails baseline). On 11,215 human navigation paths from Wikispeedia, evaluation metrics divide into two clusters that disagree with each other: alignment-based metrics (length-normalised DTW and per-step DTW similarity) favour methods that produce long alignment-rich paths, while set-overlap metrics (Jaccard and F1) favour methods that produce shorter paths with higher article overlap. On LLM-judged coherence for Cuban news storylines, evaluated under a 12-method × 5-seed × 30-endpoint-pair × 2-judge design (Claude Sonnet 4.5 and GPT-4o, both at $T=0$ via structured tool calling), the cluster-aware classical baseline is the top method in terms of mean overall coherence; the 5-method quantum-kernel pool and the 7-method classical-kernel pool on matched projection input show no significant differences after Holm correction. Cross-task analysis reveals that LLM coherence rank correlates with alignment-cluster Wikispeedia metrics (Spearman $\rho \approx +0.70$) and anti-correlates with overlap-cluster metrics ($\rho \approx -0.62$). A closed-form theoretical analysis shows that the depth-1 R_Y+CNOT-ring kernel reduces to a classical product-of-cosines kernel order equivalent to RBF, explaining the absence of empirical separation at low depth; deeper encoders break the cancellation but exponentially concentrate kernel values, eroding inter-pair distinguishability. Our results characterise quantum coherence kernels as competitive with classical kernels on the same projected input rather than decisively superior, with the cluster-aware classical baseline retaining a modest advantage attributable to its explicit topical structure. Full article

Accurate ship trajectory prediction is vital for intelligent maritime traffic management, yet conventional hybrid models often fail to balance local feature extraction, long-term dependency capture, and flexible feature weighting when processing AIS data. This paper proposes an MSC-BiLSTM-ATTENTION model that integrates trajectory clustering and an adaptive attention mechanism into a unified framework. Its fundamental advance over existing incremental hybrid architectures is twofold. First, a K-means clustering step groups trajectories with similar motion patterns before model training, effectively reducing the impact of data heterogeneity on prediction accuracy. Second, the deep learning backbone synergizes multi-scale convolution (MSC)—which captures local features at multiple temporal granularities via parallel kernels—with a bidirectional LSTM (BiLSTM) for forward–backward dependency learning, and an adaptive self-attention mechanism that dynamically optimizes feature weights to amplify critical navigation information. Extensive experiments on AIS data from the Gulf of Mexico and the U.S. Atlantic Coast, covering four seasons, benchmark the model against attention-enhanced architectures including Transformer, CNN-BiLSTM-ATTENTION, and DenseNet-BiGRU-ATTENTION across two distinct regions. The proposed model achieves significant improvements in predicting longitude, latitude, speed over ground, and course over ground, reducing MAE by over 76.9% and RMSE by over 65.3% compared with the strongest baseline. Ablation studies confirm that the synergy of all three modules is essential. The results demonstrate the model’s effectiveness and its practical value for intelligent maritime supervision, navigation risk warning, and waterborne traffic management. Full article