Search Results (376)

Machine learning combined with multispectral remote sensing provides an efficient and non-destructive approach for monitoring fruit tree nutrient status. However, conventional machine learning methods have limited ability to capture complex spectral-feature interactions, while deep neural networks often suffer from overfitting under small-sample field conditions. This study developed a UAV-based multispectral framework for estimating leaf nitrogen (N), phosphorus (P), and potassium (K) contents in a commercial pear orchard. Five-band spectral reflectance data were used to generate 21 vegetation indices (VIs), and Pearson correlation analysis and recursive feature elimination (RFE) were applied for feature optimization. A Deep Forest (DF)-based framework and a gradient boosting enhanced variant (GB-DF) were developed and compared with convolutional neural networks (CNN), support vector regression (SVR), random forest (RF), and gradient boosting decision tree (GBDT). Results showed that GB-DF achieved the best performance for N, P, and K estimation, with R² values of 0.6959, 0.7535, and 0.7216, and RMSE values of 0.7877, 0.1280, and 0.6059 g/kg, respectively. Moreover, GB-DF required only 6.35 s for training, being approximately twice as fast as GBDT and 5.7 times faster than CNN. The proposed GB-DF framework shows potential to improve nutrient estimation accuracy under limited-sample conditions and provides an effective solution for UAV-based nutrient mapping and variable-rate fertilization in precision orchards. Full article

In early-stage projects, incomplete geological data can affect the selection of retaining structures for deep excavations. This paper proposes an adaptive two-layer Case-Based Reasoning (CBR) framework to address this issue. First, a case database was constructed; feature interaction terms were incorporated into the similarity measure to capture the nonlinear coupling among geological parameters, and a Gradient Boosted Decision Tree (GBDT) was employed to achieve an objective, data-driven allocation of feature weights. The first layer implements a gating mechanism based on logical conjunction and adaptive tolerance thresholds, which automatically switches between the two inference paths when information is missing, preventing invalid hard matches; the second layer applies K-means++ clustering and local inductive reasoning to mitigate the biases caused by data sparsity. Experiments demonstrate that, under conditions of parameter incompleteness and noise interference, the method’s Top-3 recommendation accuracy significantly outperforms traditional models and machine learning baseline models and exhibits strong resistance to interference, providing a solid methodological foundation for retaining structure selection in complex data scenarios. Full article

Assessing visual fatigue is crucial in eye-controlled interaction. Traditional methods are either overly subjective or rely on highly invasive, costly equipment and complex procedures that require expert supervision. This study proposes a machine-learning-based approach for visual fatigue assessment. Data collection employs non-intrusive, easily monitored eye-tracking to capture ocular eye movement data and task performance data, while subjective questionnaires label fatigue states. For feature selection, participant-level Wilcoxon signed-rank tests with Benjamini–Hochberg FDR correction were used to identify fatigue-related indicators, and a redundancy-removal step based on Spearman correlation yielded a final set of six non-redundant features. For the assessment method, we introduced a gradient boosting decision tree–logistic regression (GBDT-LR) model whose hyperparameters are optimized via Bayesian optimization. All models were evaluated under a unified 5-fold stratified cross-validation framework with within-fold standardization and nested hyperparameter tuning. Results indicate that this model can effectively predict the state of visual fatigue. Compared with the performance of five other models—gradient boosting decision tree (GBDT), logistic regression (LR), support vector machine (SVM), random forest (RF), and RF-SVM—the proposed GBDT-LR model achieved an assessment accuracy of 89.79%, demonstrating strong predictive performance. This study provides an effective method for predicting visual fatigue in eye-controlled interaction, laying a research foundation for optimizing the user experience of eye-controlled interaction and promoting the sustainable development of eye-control technology. Full article

Gradient-boosted decision trees (GBDTs) are among the leading methods for tabular data; however, their comparative robustness to corrupted labels remains unclear. This study resolves that uncertainty with a controlled benchmark protocol in which library identity is the only free variable, so that an observed difference can be attributed to the implementation rather than to tuning, data splits, or noise realizations, and in which capacity, noise floor, and dataset-selection confound checks are mandatory before any ranking is reported. We benchmarked Extreme Gradient Boosting (XGBoost), Light Gradient-Boosting Machine (LightGBM), and Categorical Boosting (CatBoost) under symmetric, asymmetric pair-flip, and instance-dependent label noise across 15 datasets from the OpenML Curated Classification benchmark suite 2018 (OpenML-CC18), four noise rates, and three random seeds. The clean data-tuned configurations were fixed across the noise conditions. Predictive performance is the macro-averaged F1 score (macro-F1) on a clean test set, and degradation is the absolute drop in macro-F1 relative to each library’s own clean-data score on the same dataset, so the recoveries reported below are absolute percentage points of macro-F1. The ranking depends on both the noise rate and the noise model: no library separated at a 10% rate under symmetric or asymmetric noise; separation emerged from 20% under symmetric noise and only at 40% under asymmetric noise; and under instance-dependent noise, it weakened as the rate rose. At 40% symmetric and asymmetric noise, CatBoost showed significantly lower per-dataset degradation than LightGBM (Friedman tests, both $p<0.001$), with mean ranks of 1.20 versus 2.47 and 1.20 versus 2.67, respectively, and with XGBoost intermediate in both cases (2.33 and 2.13) and separable from CatBoost under symmetric noise only. At 40% instance-dependent noise, the ranking disappeared: no significant library ranking was detected ($p=0.63$), and performance gaps were smaller than twice the pooled seed standard deviation, which shows that GBDT robustness conclusions are noise-model-dependent. Rankings also varied by evaluation dimension: LightGBM had the most stable feature importances in observed means, calibration rankings depended on the noise model, and CatBoost’s training loss was the strongest mislabel-detection signal. Under symmetric noise only, small-loss reweighting significantly improved all three libraries, while early stopping recovered up to 8.5 pp for LightGBM; mitigations were not evaluated under asymmetric or instance-dependent noise. For practitioners, this means the library and the mitigation should be chosen together with the noise process that is expected: prefer CatBoost when label-conditional noise is likely and accuracy is the objective; expect no library to buy robustness under feature-dependent noise at high rates; apply early stopping or small-loss reweighting under symmetric noise, where both give significant gains and early stopping helps LightGBM most; and select on calibration, mislabel detectability, or importance stability instead when the deployment depends on those, because the ranking differs by dimension. Full article

Rice is a staple grain crop central to China’s food security. As the core production region of single-cropping rice, the middle and lower reaches of the Yangtze River face escalating high daytime and nighttime temperatures and compound drought–heat stress amid global warming. The accurate assessment of heat hazards is therefore pivotal for regional yield stability and disaster mitigation. Based on meteorological, remote-sensing, and soil data, together with county-level rice yield statistics from 150 major producing counties spanning 1991 to 2024 (5009 county-year calibration units), we first constructed a composite heat damage index (CHI) by integrating daytime harmful accumulated temperature (Ha), nighttime harmful accumulated temperature (HNa), and the Vegetation Health Index (VHI). We then implemented a gradient boosting decision tree (GBDT) machine learning framework in which yield loss was imposed as a physical constraint. This framework was benchmarked against convolutional neural network (CNN), random forest (RF), and support vector machine (SVM) models, with the Shapley additive explanations (SHAP) method used for attribution analysis and an independent temporal partitioning strategy applied for model validation. The results indicate the following: (1) compared to the single daytime heat damage index, the CHI elevated the yield correlation coefficient from 0.52 to 0.63; (2) with yield constraint calibration, the model attained a balanced accuracy of 92.6% and 94.0% consistency with historical disaster records; (3) regional heat hazard presents a spatial pattern of “high in inland areas and low in coastal areas,” with the heading–flowering stage as the critical sensitive period; and (4) high nighttime temperature accounts for approximately 20% of the model’s relative importance, with higher discriminative sensitivity for high-grade hazards, while the amplifying effect of water deficit on heat stress maintains a stable relative importance of around 16%. In this study, the coupled optimization of traditional assessment paradigms and data-driven approaches is achieved, providing a methodological reference for refined growth stage–specific heat hazard assessment. Its cross-regional portability and independent predictive validity require further validation. Full article

Foam concrete is a lightweight material characterized by low density and moderate mechanical strength, which can be combined with aluminum alloys to form a novel type of column. Such composite members enable rapid assembly, disassembly, and functional reconfiguration in prefabricated structures. However, research on the axial compressive performance of this new column system remains limited. This study investigates the axial behavior of aluminum alloy-foam concrete short columns through a combination of numerical simulation, theoretical analysis, and machine learning prediction enhanced by the SHAP (SHapley Additive exPlanations) interpretability method. A three-dimensional finite element model was developed in ABAQUS to examine the effects of frame thickness, foam concrete strength, and section dimension on load-bearing capacity. The results indicate that the column sectional dimensions have a significant influence on the axial compressive capacity. The foam concrete strength and frame thickness have relatively smaller effects. In addition, the frame thickness can effectively restrain lateral deformation and delay buckling. Based on the confinement mechanism, polynomial fitting, Mander’s model, and a composite column formulation were proposed for axial capacity prediction. Furthermore, eleven machine learning models were trained on 64 simulation datasets 64 independent computational experiments, among which the Gradient Boosting Decision Tree (GBDT) demonstrated the best performance (R² = 0.9984, MAE = 5.97, RMSE = 7.51). SHAP analysis further revealed the relative contributions of key features, showing that section dimension is the most influential parameter, followed by foam concrete strength, while frame thickness contributes the least. These findings not only enhance the theoretical understanding of the load-transfer mechanism of columns but also provide reliable predictive models and analytical formulations for their application in lightweight prefabricated structures. Full article

The Jianghan Plain is one of the important freshwater aquaculture regions in China. Accurate information on the spatial distribution and spatiotemporal dynamics of aquaculture ponds is essential for regional ecological management. However, large-scale and accurate identification remains challenging for aquaculture ponds because they have spectral and seasonal hydrological characteristics similar to those of rice fields, rivers, canals, and lakes. In this study, we developed a framework for mapping inland aquaculture ponds using multi-source remote sensing data from Sentinel-2, Sentinel-1, and PlanetScope. The framework integrated elevation-zoned Otsu thresholding for candidate water extraction, phenological features for rice field removal, and object classification based on a Gradient Boosting Decision Tree (GBDT) model using 12 shape and spatial-context features. It was applied to identify aquaculture ponds in the Jianghan Plain from 2016 to 2025. Overall accuracy exceeded 93%, and the F1-score exceeded 0.93 in all validations. Results from different sensors also showed high spatial consistency. In 2025, aquaculture ponds covered 2365.17 km² in the Jianghan Plain and were mainly concentrated in the central and eastern parts of the plain, especially near Honghu Lake and along the Yangtze and Hanjiang river meanders. Over the ten-year period, the aquaculture pond area fluctuated between 2033.58 and 2383.52 km², showing an initial decline followed by recovery. Lost aquaculture ponds were mainly located around lakes, while newly added ponds were mostly distributed along the margins of existing clusters. The decade dataset generated in this study can support freshwater aquaculture management and wetland conservation. Full article

Effective transmission of monetary policy serves as the foundational institutional guarantee for sustaining macroeconomic stability, smoothing cyclical economic fluctuations, and promoting high-quality development. Nevertheless, the underlying determinants driving the time-varying effectiveness of China’s quantity-based monetary policy have not been systematically and empirically delineated in the prevailing literature. This paper first constructs a precise measurement indicator for the effectiveness of China’s quantity-based monetary policy from the output-transmission dimension, which is defined as the response of domestic real output (excluding the contribution of net exports) to orthogonalized exogenous M2 growth shocks. On this basis, the gradient-boosting decision tree (GBDT) model is integrated with the Shapley Additive Explanations (SHAP) framework to quantitatively identify the core determinants that govern the policy effectiveness across different economic cycles and structural transformation stages. The estimation results document clear stage-wise heterogeneity in the drivers of China’s quantity-based monetary policy effectiveness: population-aging and macroeconomic-policy indicators stand out as the dominant explanatory factors over 2002–2008, while economic-structure indicators assume the leading role in shaping policy effectiveness during 2009–2015. The 2016–2022 period is further characterized by the joint dominance of demographic aging and economic-structure dimensions. Within this latest phase, the old-age dependency ratio, real–virtual economy structural misalignment, and distorted aggregate supply configuration exert statistically significant negative marginal contributions to the model-predicted effectiveness of monetary policy, whereas the total fertility rate and potential output growth rate yield positive and economically meaningful contributions. Full article

Understanding the environmental fate of microplastics (MPs) in agricultural soils remains a major challenge, particularly under field conditions where soil structure and hydraulic processes jointly regulate particle transport and retention. This study investigated whether hydro-physical soil functioning can explain the distribution and accumulation of MPs in pistachio orchard soils from a semi-arid region of southeastern Türkiye. A total of 42 soil samples were analyzed for MP abundance, size distribution, and morphology, together with key hydro-physical properties including texture, porosity, bulk density, aggregate stability, organic matter content, and soil water retention characteristics. To identify the dominant controls on MP occurrence, explainable machine learning approaches combining Random Forest (RF), Gradient Boosting Decision Trees (GBDT), and SHAP (SHapley Additive exPlanations) analysis were employed. Microplastic abundance differed among management systems. Former landfill or construction sites represented the largest proportion of the total recorded microplastic abundance (40.9%), followed by conventionally managed (25.2%), manure-amended (24.5%), and sewage-sludge-amended orchards (9.4%). Median microplastic abundances were 1433, 667, 4633, and 633 particles kg⁻¹ soil, respectively. Fine-sized MPs constituted the dominant particle fraction and exhibited strong associations with pore-system characteristics, indicating that pore-size compatibility governs their retention and mobility within the soil matrix. Morphology-specific analyses further revealed contrasting relationships between soil hydro-physical properties and individual MP forms, suggesting distinct retention pathways for granules, films, fragments, and fibers. Explainable AI analysis identified organic matter, silt content, bulk density, and water retention characteristics as the most influential predictors of MP occurrence. Among the tested models, RF demonstrated superior predictive robustness and generalization capacity. The findings demonstrate that hydro-physical soil functioning plays a central role in determining microplastic fate in agricultural soils and highlight the value of interpretable machine learning frameworks for uncovering the mechanisms underlying contaminant retention and redistribution. Integrating soil structural indicators with explainable artificial intelligence offers a promising pathway for improving microplastic risk assessment in agroecosystems. Full article

Artificial intelligence (AI) systems for intensive care units (ICUs) must support early risk prediction while producing explanations that clinicians can inspect, question, and relate to physiological reasoning. Deep neural networks (DNNs) can learn complex temporal patterns from electronic health records (EHRs), but their internal representations are often difficult to translate into clinically actionable explanations. Gradient-boosted decision trees (GBDTs) offer more transparent decision rules, yet they may not capture the full temporal and nonlinear structure of high-dimensional ICU data. This paper presents a causal-pathway-guided DNN–GBDT distillation framework for interpretable ICU decision support. The framework first estimates a directed acyclic graph (DAG), denoted by $\mathcal{G}$, from multivariate ICU time-series data and then uses the graph to guide representation learning in a DNN teacher model through causal gating. The learned teacher is distilled into a GBDT student model using soft predictive targets and a causal attribution-guided split-selection procedure, so that the final model approximates the teacher predictions while prioritizing tree splits aligned with plausible physiological pathways. Experiments using Medical Information Mart for Intensive Care IV (MIMIC-IV) data evaluate sepsis onset and in-hospital mortality prediction through discrimination, precision–recall performance, calibration-oriented reporting, causal consistency, and clinical utility indicators. The proposed causal-aware distilled GBDT achieves stronger predictive performance than conventional interpretable baselines and substantially higher causal consistency than black-box temporal models. The results suggest that causal structure can serve as an inductive bias for converting complex temporal prediction into interpretable rule-based clinical reasoning. The paper also discusses limitations related to observational causal discovery, unmeasured confounding, temporal stationarity, and clinical deployment, following recent reporting expectations for AI-based clinical prediction models. Full article

Efficient water vapor removal is important for both building humidity control and industrial gas dehydration, where operating conditions may span broader temperature and pressure ranges. Driven by a pressure gradient, membrane-based dehumidification has emerged as an energy-efficient alternative, employing polymeric composite membrane materials to achieve effective moisture separation. However, traditional development of such membranes remains heavily reliant on inefficient trial-and-error approaches. To overcome this limitation, this study employs machine learning to directly predict the relationships between physicochemical structure, operational conditions, and water vapor permeation performance of composite membrane materials. A dataset comprising 138 experimental samples from 26 published studies was compiled, featuring five input features: selective layer thickness, operating temperature, feed pressure, relative humidity, and a newly proposed hydrophilicity score based on functional group composition. Among six machine learning models evaluated, the Gradient Boosting Decision Tree (GBDT) achieved superior predictive performance, yielding a test R² of 0.912. SHAP analysis identified selective layer thickness as the dominant descriptor, followed by feed pressure, hydrophilicity score, operating temperature, and relative humidity, contributing 34.4%, 26.5%, 15.8%, 11.9%, and 11.5% to the model predictions, respectively. Within the investigated parameter space, a genetic algorithm integrated with the GBDT model identified a permeability-oriented parameter combination (18.25 μm thickness, 111.43 °C, 0.94 bar, 52.02%RH, and a hydrophilicity score of 5), achieving a predicted permeability of 136,418 Barrer. The framework offers a transferable strategy for accelerating the rational design of advanced membrane materials, significantly reducing the need for exhaustive experimental screening. Full article

Differential surface settlement is a critical trigger for engineering disasters, with buried ground fissure activity serving as one of the core factors driving such subsidence. In this study, four groups of physical model tests were conducted. High-precision laser displacement meters were used to monitor real-time settlement at varying distances from the ground fissure. Based on this, the surface displacement evolution patterns of both loess layer structures and interbedded sand–soil structures under slow (0.01 m/h) and rapid (2 m/h) ground fissure activities were summarized. To evaluate the predictive performance of different models for time- and space-dependent nonlinear displacement evolution, five methods, namely polynomial regression, support vector regression (SVR), multilayer perceptron (MLP), random forest (RF), and gradient boosting decision tree (GBDT), were constructed and compared using the experimental displacement dataset. The physical model test results indicate that the interbedded sand–soil structure tends to suppress localized crack propagation while expanding the affected zone range, whereas the loess layer is prone to near-field deformation localization. The activity rate exerts a pronounced influence on fracture propagation within the loess layer, while its influence on the deformation of the interbedded sand–soil structure is relatively limited under the present test conditions. The model comparison results show that a cubic polynomial can effectively describe the displacement evolution pattern under slow and homogeneous conditions. MLP exhibited the most stable performance among the four machine learning algorithms, while SVR, GBDT, and RF provided complementary information for interpreting continuous displacement evolution and local nonlinear displacement variations. Overall, under the present physical model test conditions, the machine learning models provide a useful data-driven basis for quantitatively characterizing displacement evolution trends and interpreting the development of ground fissure-affected zones. Based on the experimental results—scaled up by a similarity ratio of 20:1 and incorporated with a safety factor of 1.1~1.3—the preliminary reference engineering avoidance distance in the ground fissure-affected zone is 20~24 m for the hanging wall and 14~16 m for the footwall, which are broadly consistent with the current codes and regulations. The findings provide controlled experimental evidence and a quantitative reference for ground fissure hazard assessment and engineering protection. Full article

The compressive strength of ultra-high-performance concrete (UHPC) is jointly influenced by multiple factors, including material composition, mixture proportion parameters, and curing regime. Conventional empirical methods are therefore insufficient to accurately characterize the highly nonlinear relationships involved. To improve the prediction accuracy of UHPC compressive strength and to achieve mixture proportion optimization that simultaneously considers mechanical performance, economic efficiency, and environmental impact, this study developed random forest (RF), artificial neural network (ANN), gradient boosting decision tree (GBDT), and extreme gradient boosting (XGBoost) models based on 810 publicly available UHPC experimental datasets. Model performance was evaluated using R², RMSE, MAE, and MAPE. To enhance the robustness of model validation, repeated K-fold cross-validation, sensitivity analysis with different random seed splits, and benchmark model comparisons were further introduced. The results indicate that the XGBoost model achieved superior predictive performance on both the test set and robustness validation, with test-set R², RMSE, MAE, and MAPE values of 0.9604, 7.77, 5.58, and 4.80, respectively. The model was further interpreted using SHAP, PDP, and ICE methods, and the results revealed that curing age, fiber content, silica fume content, and water-to-binder ratio were important variables affecting the compressive strength of UHPC. Furthermore, XGBoost was used as a surrogate model and coupled with NSGA-II and TOPSIS methods for multi-objective optimization. Under the constraints of compressive strength, water-to-binder ratio, superplasticizer-to-binder ratio, and absolute volume, a computationally recommended UHPC mixture proportion balancing strength, cost, and carbon emissions was obtained. This study provides a reproducible machine-learning-assisted approach for UHPC compressive strength prediction and low-carbon, cost-effective mixture proportion design. Full article

The relationship between urban waterfront spaces and health is widely recognized. However, the relative importance and nonlinear associations of specific attributes with residents’ physical health remain unclear. Drawing on survey data from 375 users across seven waterfront parks in Changchun, this study uses a Gradient Boosting Decision Tree model to examine how 37 perceived attributes of waterfront recreation spaces contribute to self-assessments of physical health, focusing on their relative contributions and nonlinear associations. The aim is to provide exploratory evidence for the empirical accumulation of the relationship between waterfront spaces and health. The results show that waterfront walkways and plaza lighting had high relative importance in the prediction model, at 8.8% and 8.2%, respectively. These findings suggest that spatial connectivity and around-the-clock accessibility may play a fundamental role in creating health-supportive environments. Plant landscapes, seating, pet-friendly facilities, waterside platforms, and open lawns for public access constitute secondary dimensions, with relative contributions ranging from 3.1% to 5.4%. The partial dependence plots of most attributes exhibit nonlinear associations. The pet-friendly facilities are presented in a two-step pattern in the partial dependence plot. This study proposes a conceptual reference framework for planning and offers exploratory proposals for waterfront recreation spaces. Full article

Ozone (O₃), as a key secondary pollutant, exhibits pronounced spatiotemporal heterogeneity, posing significant challenges to coordinated regional air pollution control. However, systematic understanding of high-resolution O₃ spatial inversion and its driving mechanisms in arid urban agglomerations remains limited. In this study, the Tianshan North Slope Urban Agglomeration (TNSUA) was selected as the study area, and a multi-model comparative framework was established to comprehensively evaluate the O₃ inversion performance of 16 machine learning and deep learning models, including Extreme Gradient Boosting (XGBoost), Random Forest (RF), Extremely Randomized Trees (ET), and Gradient Boosting Decision Tree (GBDT). Based on the optimal model performance, high-precision daily O₃ spatial reconstruction for the year 2023 was achieved across the study region. The contributions of individual driving factors and their nonlinear response relationships were quantitatively interpreted using Shapley Additive Explanations (SHAP). Furthermore, a backward trajectory model combined with the Weighted Potential Source Contribution Function (WPSCF) and Weighted Concentration Weighted Trajectory (WCWT) methods was employed to identify potential source regions and transport pathways of O₃. The results indicate that: (1) The XGBoost model exhibited the best performance (R² = 0.93, RPD > 3). The reconstructed results reveal that high O₃ concentrations in 2023 were primarily distributed in southern Urumqi, southern Changji, and southern Tacheng, with southern Urumqi identified as the most prominent hotspot. (2) The spatial variability of O₃ was predominantly driven by downward shortwave radiation (DSR) and air temperature (TEM), both of which showed significant nonlinear responses and threshold effects on O₃ formation. (3) Source apportionment analysis indicates that westerly transport serves as a major exogenous contribution pathway, with potential source regions mainly located in the surrounding areas of the northern Tianshan slope as well as Central Asia, particularly eastern Kazakhstan and northern Kyrgyzstan. This study systematically elucidates the formation mechanisms of O₃ pollution in arid urban agglomerations from three aspects—high-precision inversion, driving mechanism analysis, and cross-regional transport identification—thereby providing a scientific basis for precise air pollution control strategies. Full article