Search Results (459)

Musculoskeletal injury remains a major welfare and performance concern in Thoroughbred racing, and biomarkers for early risk stratification are of scientific interest. This multicentre observational cohort study characterised serum concentrations of two bone turnover markers, osteocalcin/BGLAP (OC) and C-terminal telopeptide of type I collagen (CTX-I), in 1359 fit-to-race Thoroughbred racehorses sampled across New South Wales, Victoria, and Western Australia. Biomarker distributions, population-wide percentile ranges, demographic and training-related associations, and short-term soundness outcomes were evaluated using nonparametric methods. Both markers were positively skewed, with population-wide percentile ranges (2.5–97.5 percentile) of 0.13–9.44 ng/mL for CTX-I and 0.02–7.25 ng/mL for OC; CTX-I varied substantially by jurisdiction, so this pooled range should not be interpreted as a clinical reference interval for individual horses. CTX-I varied significantly by age, jurisdiction, venue, and training surface, with markedly higher concentrations in New South Wales horses and lower concentrations associated with polytrack training, but did not predict lameness outcomes. OC concentrations were significantly higher in horses classified as lame at sampling and showed modest discriminatory ability for subsequent lameness in univariate analysis, with the strongest performance for persistent lameness at both 7 and 28 days; this association attenuated after adjustment for age, sex, state, and training surface. At an unadjusted operational threshold of approximately 1.24 ng/mL, OC achieved high negative predictive value for persistent lameness. These findings suggest OC may have scientific value as a potential rule-out screening marker for short-term soundness, warranting further investigation, while CTX-I appears more informative for characterising population-level skeletal turnover variation. Full article

Recommendation systems in public libraries rely on loan data skewed toward past popularity, making it difficult for unborrowed and newly published titles to reach users. Hence, we propose and evaluate a machine learning-based approach predicting library material popularity using open loan data from South Korean public libraries. Three feature sets were constructed: word2vec-based title embeddings (F₁), borrower demographic features (gender and age group; F₂), and topic features from the Korean Decimal Classification (KDC) main class (F₃). Seven machine learning models were evaluated using title-level grouped cross-validation, and XGBoost was selected as the best-performing model. Using this model, the effect and marginal contribution of the feature sets were examined via pairwise t-tests on title-level grouped cross-validation repeated 30 times. Consequently, the full feature set F outperformed all two-feature-set combinations, and F₃ emerged as a key feature set. The feature sets were consistently ranked F₃ > F₂ > F₁ in both predictive performance and model fitness. A cold-start evaluation confirmed near-identical performance for entirely unseen titles. Thus, library material popularity can be predicted using only publicly available information, suggesting the feasibility of a privacy-preserving approach to informing library material recommendations, relevant to library use, digital inclusion, and social justice research. Full article

The Rayleigh–Logarithmic distribution provides a useful framework for modelling lifetime data by combining continuous lifetime variability with a logarithmic compounding mechanism. This study introduces a three-parameter Alpha Power Rayleigh–Logarithmic (APRL) distribution by applying the Alpha Power transformation to the classical Rayleigh–Logarithmic model. The additional transformation parameter allows the distributional shape, skewness, tail behaviour, and rate of increase in the hazard function to be adjusted while retaining the underlying structure of the baseline model. Several mathematical and reliability properties of the APRL distribution are derived, including the probability density and cumulative distribution functions, survival and hazard rate functions, quantile function, moments, order statistics, and mean residual life function. Model parameters are estimated by maximum likelihood using a multiple-start numerical optimization procedure, and the finite-sample performance of the estimators is investigated through Monte Carlo simulations under different parameter configurations and sample sizes. The simulation results show that estimation accuracy generally improves with increasing sample size, as reflected by decreasing bias, MSE, and RMSE, although estimation of the transformation parameter may exhibit greater variability for more extreme parameter settings. The practical performance of the APRL distribution is examined using the Aircraft Windshield Failure Times and Breaking Stress of Carbon Fibres datasets. Model comparisons based on information criteria, bootstrap-based goodness-of-fit assessment, and graphical diagnostics show that the APRL distribution provides competitive fits relative to several established lifetime distributions. In addition, mean time to failure and mean residual life analyses illustrate the practical interpretation of the reliability measures derived for the proposed model. Overall, the results support the APRL distribution as a useful alternative for the statistical analysis of lifetime and reliability data. Full article

Reliable rainfall regionalization underpins hydrological design and the transfer of rainfall information to ungauged or short-record sites, yet it is especially difficult in arid regions with strong spatial and temporal variability and short, uneven records. This study develops a multivariate, stability-validated framework for classifying rain gauges into candidate homogeneous rainfall regions across Saudi Arabia. A national database of 274 stations was screened for 2015–2023, and 202 stations were retained. Ten descriptors representing annual rainfall, interannual variability, L-skewness of annual maxima, within-day rainfall concentration, short-duration intensity, elevation, and distance from the coast were constructed from 5 min records; correlated concentration and intensity descriptors were compressed by block-wise principal component analysis into a seven-variable, robustly scaled feature matrix. Partitions from K = 2 to 10 were evaluated using internal-validity indices, the gap statistic, minimum cluster size, repeated-subsample stability, cross-algorithm agreement, and sensitivity to alternative feature representations. The diagnostics supported several low-order structures: the four-region K-means solution showed the highest subsample stability (median adjusted Rand index: 0.977) and best consensus rank, whereas Ward clustering and Gaussian mixture modeling favored a parsimonious three-region solution. At four regions, cross-algorithm agreement was moderate (adjusted Rand index: 0.526–0.663) and the partition was strongly reproduced under the robust original-variable and global principal component analysis-(PCA)PCA representations (0.804). The regions comprised a widespread arid interior, a small near-coastal group, a wetter western–southwestern group, and a coastal-foothill group, with distance from the coast, elevation, and short-duration intensity providing the strongest contrasts. The framework offers a reproducible basis for regional rainfall-frequency analysis, station pooling, and hydrological transfer, pending verified completeness data and formal homogeneity testing. Full article

Bus-mounted vision sensing provides a practical and complementary perspective for intelligent transportation systems, but reliable traffic object detection from bus front-view cameras remains challenging because elevated viewpoints induce severe scale skewness, dense interactions around bus stops and intersections, and frequent heterogeneous occlusion. To support this sensing scenario while avoiding ambiguity with previously used dataset acronyms, we construct the Bus Front-view Traffic Dataset (BFTD), a high-resolution benchmark collected from forward-facing cameras mounted on multiple buses operating on urban routes during real-world service. The BFTD contains 8131 images and 56,137 annotated instances across five traffic-participant categories, covering dense pedestrians, mixed-traffic flow, illumination variation, rain, fog, and occlusion-prone scenes. Based on the visual characteristics of bus-mounted cameras, we propose YOLO-M2LA, a local–global attention detection framework in which CBS-SPD preserves fine-grained information during early downsampling and M2LA couples multi-scale local context modeling with efficient global dependency aggregation. Extensive experiments on BFTD and public benchmarks show that the proposed framework improves detection accuracy, particularly for small and visually crowded traffic participants, while maintaining a practical accuracy–efficiency trade-off. Dataset statistics, condition-specific evaluation, ablation analysis, and qualitative visualization further support the effectiveness of BFTD and YOLO-M2LA for vision-based traffic sensing. The dataset and implementation are publicly available online. Full article

Estimating the finite population mean under systematic sampling becomes challenging when auxiliary information is nonlinear, skewed, or structurally complex, as conventional linear estimators often lose efficiency. This study proposes a new class of weighted hybrid estimators that combine harmonic and geometric transformations of the auxiliary variable. The proposed approach is designed to capture nonlinear relationships while handling skewed data and reducing sensitivity to extreme observations. Expressions for bias and mean squared error are derived, and optimal weights are obtained by minimizing the mean squared error. The theoretical results indicate that the proposed estimators are more efficient than traditional ratio, product, regression, and exponential-type estimators. A simulation study further confirms their improved performance across various population structures, correlation levels, and sampling fractions, with notable improvements in skewed and nonlinear settings. The proposed class provides a flexible and reliable alternative for practical applications in systematic sampling. Full article

Indigenous languages in Southern Thailand are increasingly affected by modernization, digital communication, and language contact, yet the relationship between kinship terminology, social organization, and linguistic sustainability among Indigenous Austroasiatic-speaking communities remains insufficiently explored. This study addresses three research questions: (1) how Tan’ean and Jaahaai kinship systems differ in their semantic structures and classificatory patterns; (2) how kinship terms reflect cultural worldviews, social organization, and relational values of the Manni and Mueng’ra communities; and (3) what challenges influence the transmission and sustainability of kinship-related linguistic knowledge. This study aims to comparatively analyze consanguineal and affinal kinship terminology in the Tan’ean language of the Manni ethnic group and the Jaahaai language of the Mueng’ra ethnic group, focusing on how kinship systems encode cultural worldviews, social relations, and adaptive strategies under contemporary change. Guided by Componential Analysis and Ethnolinguistic Vitality Theory, this study employed a qualitative ethnolinguistic approach. Data were collected through genealogical elicitation, in-depth interviews, and participant observation with 40 key informants from Trang, Phatthalung, Satun, Songkhla, Yala, and Narathiwat provinces. The data were analyzed through semantic feature identification and comparative interpretation of kinship structures. The findings demonstrate that, despite shared Austroasiatic heritage, Tan’ean and Jaahaai exhibit distinct kinship systems reflecting different cultural logics. In response to the first research question, the findings reveal that the Manni system resembles the Hawaiian type, emphasizing egalitarian relations, generational categories, and inclusive incorporation of affinal relatives, whereas the Mueng’ra system shows Crow–Omaha characteristics, including generational skewing, gender-specific markers, hierarchical classification, and stronger influence from dominant regional languages. Regarding the second research question, the findings demonstrate that kinship terminology functions as a cultural mechanism for organizing social relationships, identity, and intergenerational obligations, reflecting different models of Indigenous social organization. For the third research question, the study identifies language contact, changing social conditions, and reduced intergenerational transmission as challenges affecting the continuity of Indigenous kinship knowledge. Therefore, the study recommends community-based orthographies, participatory digital learning platforms, and policies supporting Indigenous linguistic rights to strengthen language revitalization, cultural continuity, and Indigenous communities’ capacity to adapt to social transformation. Full article

Asymmetric interaction matrices encode the linear coupling structure and directed interaction patterns that arise in mathematical models of complex networks across ecology, finance, and machine learning, yet their geometric structure as points on a statistical manifold has received comparatively little systematic treatment. This paper develops a rigorous information-geometric framework for the manifold ${\mathcal{M}}_n^{+}$ of real $n\times n$ interaction matrices whose symmetric part is negative definite—equivalently, the matrices satisfying the numerical stability condition $\omega \left(A\right)={\lambda }_{\max }\left(S\left(A\right)\right)<0$. The symmetric part $S\left(A\right)=(A+A^{\mathsf{T}})/2$ and the skew-symmetric part $K\left(A\right)=(A-A^{\mathsf{T}})/2$ correspond, respectively, to the metric structure and the torsion of the induced statistical manifold. We construct the natural augmented Riemannian metric g on ${\mathcal{M}}_n^{+}$ as the sum of the Fisher–Rao pullback metric through $-S(·)$ and a Frobenius term on $K(·)$, derive explicit formulae for the sectional curvature in the mixed symmetric–skew plane, and prove that the sectional curvature vanishes if and only if A is normal. The central theoretical result is a curvature-mediated stability theorem: a Fisher–Rao stability margin, derived from the precision representative of A, provides a sharp, computationally accessible certificate for the asymptotic stability of the linear dynamical system $\dot{\mathbf{x}}=A\mathbf{x}$, with the instability boundary lying at infinite Fisher–Rao distance. We further establish an information-geometric reformulation of May’s stability criterion for random ecological networks, a curvature-based covariance regularisation scheme for financial correlation matrices, and a Jacobian stability bound for deep neural networks. All the main results are illustrated with explicit $3\times 3$ and $4\times 4$ numerical examples. Full article

Clustered proportion responses often exhibit bounded support, skewness, heterogeneous dispersion, and within-cluster dependence. We propose a correlated mean–precision random-effects beta regression model that jointly represents cluster-level heterogeneity in the conditional mean and conditional precision. Its main innovation is to treat the cross-submodel random-effect correlation as a scientific estimand. The same frequentist Beta mixed model estimates and tests this correlation while allowing nonlinear adjustment and cluster-level interpretation. Penalized B-splines allow nonlinear effects in both submodels, and estimation is performed by maximizing a Laplace-approximated penalized marginal likelihood. The fitted model provides likelihood-based inference for the mean–stability association and empirical Bayes estimates for cluster ranking and quadrant classification. Among M1–M5, M5 gives the lowest average errors for conditional-mean and conditional-precision recovery and the best average AIC, cluster-level BIC, and full-data NLPD across the Monte Carlo settings considered. It is also the only model compared here that estimates and tests the latent association while retaining paired cluster effects for interpretation. All M5 fits succeeded in the enlarged 30-replication stress suite, and the five aggregate mean M5–M4 criteria favored M5 across 299 successful pairs; quadrature checks indicate where safeguards are needed under weak information. In both CDC PLACES and the representative World Bank panel of 128 eligible countries, M5 has the largest marginal likelihood, the smallest AIC and cluster-level BIC, a significant M4–M5 likelihood-ratio test, and the lowest application-specific point-prediction errors. For the World Bank data, $\widehat{\rho }=-0.500$ (profile 95% interval $-0.621$ to $-0.295$), the likelihood-ratio statistic is $17.370$ ($p=3.08\times {10}^{-5}$), and M5 has the lowest rolling-origin MSPE, RMSE, and MAE. On the combined evidence from fit, prediction, and correlation inference, M5 is the best overall model evaluated in both applications. Full article

Background: Anti-Müllerian hormone (AMH) is widely used as a marker of ovarian reserve and is strongly influenced by chronological age. Although inflammation has been implicated in ovarian aging and follicular dysfunction, whether routinely measured peripheral inflammatory markers provide additional information regarding AMH concentrations remains unclear. Objective: This study aimed to evaluate the age-adjusted associations of C-reactive protein (CRP) and the neutrophil-to-lymphocyte ratio (NLR) with serum AMH concentrations in reproductive-age women. Methods: This retrospective cross-sectional study included women aged 18–45 years with available AMH, complete blood count, and CRP measurements. After reapplying the prespecified eligibility criteria and excluding confirmed data-entry or analytical errors, 819 women were included in the final analysis. NLR was calculated from absolute neutrophil and lymphocyte counts. Because CRP and NLR showed right-skewed distributions, Box–Cox transformations were applied. Associations with Box–Cox-transformed AMH were evaluated using an age-adjusted left-censored Tobit regression model. Results: The mean age of the participants was 34.07 ± 6.94 years, and the median AMH concentration was 1.07 ng/mL (interquartile range, 0.29–2.50). Chronological age was inversely associated with AMH (β = −0.123, 95% CI: −0.135 to −0.111, p < 0.001). After adjustment for age, neither CRP (β = −0.038, 95% CI: −0.085 to 0.009, p = 0.114) nor NLR (β = −0.012, 95% CI: −0.222 to 0.198, p = 0.912) was independently associated with AMH. Conclusions: In this retrospective outpatient cohort, routine peripheral inflammatory markers did not explain additional variation in AMH beyond chronological age. These results do not exclude a potential role for local ovarian inflammation, which may not be adequately captured by peripheral CRP or NLR measurements. Full article

Cotton boll opening is an important late-season indicator for maturity assessment, defoliation timing, and harvest planning, but field-level monitoring remains challenging because visible lint progression varies spatially and temporally. This study evaluates whether UAV-derived cotton visible lint percentage, aggregated at the Sentinel-2 10 m grid scale, can be estimated using Sentinel-2 spectral bands, vegetation indices (VIs), and accumulated growing degree days (AGDD) as phenological predictors. UAV multispectral imagery was used to derive visible lint percentage through red-band thresholding and segmentation within canopy masks. The UAV-derived visible lint information was summarized within fixed Sentinel-2 10 m grid cells to generate satellite-compatible response labels. Four supervised regression models, eXtreme Gradient Boosting (XGBoost), Random Forest (RF), k-Nearest Neighbors (kNN), and Neural Network/Multilayer Perceptron (NNET/MLP), were evaluated using raw and transformed target formulations. Raw visible lint percentage produced relatively high explanatory power for tree-based models, with R² values of 0.73 for both XGBoost and RF. However, the target distribution was strongly right-skewed and dominated by low visible lint values, with a mean PCTOPEN of 4.05%Open, motivating the evaluation of target transformations to reduce target skewness while assessing their impact on predictive performance. On the original PCTOPEN scale, the raw target produced RMSE = 4.38%Open points, MAE = 2.31%Open points, and MedianAE = 0.68%Open points. The square-root transformation provided the strongest overall predictive performance, maintaining R² = 0.73 and RMSE = 4.38%Open points while reducing MAE to 2.14%Open points and MedianAE to 0.41%Open points. Stronger transformations further reduced the typical prediction errors, with MedianAE = 0.35, 0.33, and 0.31%Open points for the cube-root, fourth-root, and fifth-root transformations, respectively. However, these improvements were accompanied by progressively lower R² values (0.72, 0.71, and 0.69) and higher RMSE values (4.48, 4.58, and 4.68%Open points), indicating that stronger transformations reduced typical prediction errors at the expense of overall predictive performance. These results indicate that UAV-derived visible lint percentage can be linked with Sentinel-2 observations for satellite-scale regression modeling, but prediction uncertainty remains influenced by target skewness, mixed 10 m pixels, canopy obstruction, and limited UAV acquisition density. Additional UAV–Sentinel-2-aligned acquisitions, supporting multi-field observations, and broader validation across fields, seasons, cultivars, and production environments are needed to improve robustness and support operational cotton boll-opening tracking applications. Full article

Contemporary discussions of collective attention in digital environments often presuppose its existence without specifying the conditions under which it emerges, stabilizes, or fragments. This paper reconceptualizes collective attention not as shared mental focus or institutionally coordinated practice, but as an emergent socio-technical phenomenon grounded in the regulation of inference under uncertainty. Drawing on predictive processing, attention is defined as a structuring effort that governs how environmental information is received to support action-guiding inference. The paper develops a precision-based framework distinguishing two ideal-typical regimes: organic attention, which arises when precision remains responsive to uncertainty; and mechanistic attention, which emerges when precision is decoupled from temporal engagement density and redirected by engagement-optimizing architectures—a phenomenon conceptualized as precision hijacking. The framework is operationalized through two system-level indicators—Collective Free Energy (CFE) and Precision Alignment Index (PAI)—applied to temporal interaction data from Wikipedia talk pages. Empirical analysis of Wikipedia talk pages (N = 379,978 articles, N = 191,372 contributors) reveals a collective attention regime characterized by substantial CFE (41,649,821.60), moderate PAI (0.089), and a highly skewed precision distribution (Median = 0.000). Results demonstrate that collective attention can stabilize in minimally institutionalized digital ecologies, exhibiting bounded collective free energy and structured precision alignment. The weak and complex correlation between temporal engagement and precision (Pearson r = 0.059, Spearman ρ = −0.172) suggests that these dimensions are partially coupled but moderated by other factors. These findings show that institutionalization enhances stability but is not a necessary condition for collective attentional emergence. The paper contributes a diagnostic framework for analyzing digital attention economies and offers new resources for the governance of collective sense-making. Full article

Background: The widespread use of appearance-focused social media platforms has transformed how body-related attitudes are constructed, internalized, and evaluated, particularly among adolescent girls and young women, yet the existing literature remains heterogeneous in conceptual frameworks, populations, and outcome measures. Objective: This scoping review maps the extent, nature, and characteristics of the available evidence on the relationship between appearance-focused social media use, body image, self-esteem, and psychological wellbeing in adolescent girls and young women. Methods: Following PRISMA-ScR, PubMed, Scopus, and Web of Science were searched for quantitative, qualitative, and mixed-methods studies on female samples aged 10–30; data were charted and synthesized descriptively. Results: Fifty-eight studies (2014–2026) met the inclusion criteria. Evidence was concentrated in Australia and the United States and skewed toward emerging adults, with adolescent girls aged 10–17 examined in only nine studies. Appearance-focused engagement was consistently linked to body dissatisfaction, body surveillance, drive for thinness, and negative affect, with appearance comparison, internalization of appearance ideals, and self-objectification emerging as recurrent explanatory mechanisms. Body positive and body neutrality content emerged as the most promising protective categories. Conclusions: The review clarifies current evidence and identifies critical gaps for future research, informing clinical, educational, and preventive interventions. Full article

This paper proposes a personalized federated learning framework for appliance recognition in non-intrusive load monitoring (NILM) to address real-world data heterogeneity. Each client maintains a personalized model alongside shared global components. To decouple these components, a context-aware conditional policy module adaptively separates global and personalized information via learnable gating. The method enables collaborative training without raw data exchange and mitigates the impact of inter-client label distribution skew. We evaluate the proposed method under four federated settings: independent and identically distributed (IID), Dirichlet non-IID, house-partitioned, and leave-one-house-out. Experiments on three public datasets show strong and stable performance compared with existing federated approaches. Under Dirichlet skew ($\alpha =0.1$), our method achieves an accuracy of $93.8\pm 0.9\%$ on PLAID, $92.3\pm 1.4\%$ on WHITED, and $96.6\pm 1.5\%$ on COOLL. In the leave-one-house-out setting, it attains $80.3\pm 2.0\%$ on PLAID. These results demonstrate the effectiveness of the proposed method across challenging non-IID scenarios. Full article

Understanding how species richness is distributed across space is a central goal of macroecology and biogeography. The gecko genus Hemidactylus ranks among the most speciose and broadly distributed of all squamate genera, with species spanning tropical and subtropical environments on most continents. Yet the large-scale structure of its diversity has rarely been examined at a global scale. Here we assembled georeferenced occurrence data for the genus from the Global Biodiversity Information Facility (GBIF Access data is 14 March 2026) and, following cleaning and quality control, retained 143,858 records covering 157 currently recognized species (approximately 79% of the genus) to characterize broad-scale biodiversity patterns. Our analyses addressed the latitudinal diversity gradient, the distribution of species range sizes, continental richness, spatial clustering of diversity, and regional beta diversity. Richness declined steadily from the tropics toward the poles, producing a pronounced latitudinal diversity gradient (Pearson r = −0.71, p < 0.001). A generalized additive model captured a strong non-linear association between latitude and richness (deviance explained = 98.1%), consistent with tropical regions acting as principal centers of diversity. Range-size distributions were markedly right-skewed, with the majority of species restricted to comparatively small areas; however, these estimates co-varied strongly with per-species sampling effort (r = 0.90) and warrant cautious interpretation. We found only a weak, marginally significant association between range size and latitudinal midpoint, offering little support for Rapoport’s rule in the genus. At the continental scale, Hemidactylus diversity was overwhelmingly concentrated in Afro-Asian tropical regions (Asia, 93 species; Africa, 66), whereas Europe and Oceania supported few species (four each). Spatial analyses indicated significant autocorrelation in richness (Moran’s I = 0.56, p < 0.001) and well-defined diversity hotspots in East Africa, the Indian subcontinent, and Southeast Asia, and regional beta diversity was high (Jaccard dissimilarity up to 0.97), reflecting pronounced faunal turnover among continents. Full article