Search Results (260)

Long landings can reduce runway utilization and increase the probability of runway incursions and excursions. Previous studies on long landings often lacked support from actual operational data and primarily relied on event-triggering logic established by airlines for parameter exceedance detection and retrospective analysis. In response, a comprehensive risk prediction framework for aircraft long landings, supported by Quick Access Recorder (QAR) data, was constructed. The framework includes a data analysis pipeline, a sequence prediction model, and performance evaluation metrics for accident warning efficiency. Specifically, approximately 3 million rows of real QAR data were collected, and reasonable landing intervals were extracted based on pilots’ correct landing sightlines, attention allocation, and actual visual scenarios at departure heights. Gradient Boosting Decision Trees (GBDT) were employed to develop a method for extracting landing interval feature data, based on monitored parameters and ranges of landing distance. Additionally, the GBDT-Informer long-sequence time series prediction model was developed to forecast landing distance, accompanied by the construction of effective metrics for evaluating prediction performance. The results indicate that the GBDT-Informer model effectively models the temporal dimensions of landing intervals, accurately predicting ground speed (GS), radio altitude (RALT), and landing distance sequences. Compared to other prediction models, the GBDT-Informer model consistently achieved the smallest RMSE, MAE, and MAPE values, demonstrating high prediction accuracy. This predictive framework allows for the analysis of the coupling relationships among multiple parameters in flight data and their interrelations with exceedance anomalies. The findings can be applied in actual flight landings to promptly assess whether landing distances exceed limits, providing quick references for flight crews during landing or go-around decisions, thereby enhancing operational safety margins during the landing phase. Full article

To address the negative sample selection bias and limited interpretability of traditional debris-flow event susceptibility models, this study proposes a framework that enhances generalization by integrating negative sample screening via a stacking ensemble model with an interpretable random forest. Using Wenchuan County, Sichuan Province, as the study area, 19 influencing factors were selected, encompassing topographic, geological, environmental, and anthropogenic variables. First, a stacking ensemble—comprising logistic regression (LR), decision tree (DT), gradient boosting decision tree (GBDT), and random forest (RF)—was employed as a preliminary classifier to identify very low-susceptibility areas as reliable negative samples, achieving a balanced 1:1 ratio of positive to negative instances. Subsequently, a stacking–random forest model (Stacking-RF) was trained for susceptibility zonation, and SHAP (Shapley additive explanations) was applied to quantify each factor’s contribution. The results show that: (1) the stacking ensemble achieved a test-set AUC (area under the receiver operating characteristic curve) of 0.9044, confirming its effectiveness in screening dependable negative samples; (2) the random forest model attained a test-set AUC of 0.9931, with very high-susceptibility zones—covering 15.86% of the study area—encompassing 92.3% of historical debris-flow events; (3) SHAP analysis identified the distance to a road and point-of-interest (POI) kernel density as the primary drivers of debris-flow susceptibility. The method quantified nonlinear impact thresholds, revealing significant susceptibility increases when road distance was less than 500 m or POI kernel density ranged between 50 and 200 units/km²; and (4) cross-regional validation in Qingchuan County demonstrated that the proposed model improved the capture rate for high/very high susceptibility areas by 48.86%, improving it from 4.55% to 53.41%, with a site density of 0.0469 events/km² in very high-susceptibility zones. Overall, this framework offers a high-precision and interpretable debris-flow risk management tool, highlights the substantial influence of anthropogenic factors such as roads and land development, and introduces a “negative-sample screening with cross-regional generalization” strategy to support land-use planning and disaster prevention in mountainous regions. Full article

Accurate prediction of geological hazard susceptibility forms the foundation of effective risk management, yet small-sample constraints often limit model generalization. In order to address this issue, this study applied an ensemble method based on predictive symmetry quantification, using Mount Tai, China, as a test case. Thirteen influencing factors were integrated using six machine learning algorithms—Logistic Regression (LR), Multilayer Perceptron (MLP), Random Forest (RF), Gradient Boosting Decision Tree (GBDT), eXtreme Gradient Boosting (XGB), and Support Vector Machine (SVM)—trained on 34 hazard sites. Symmetry breaking in model outputs was quantified, and XGB and MLP, which showed the lowest correlation (0.59), were selected for dynamic weighted integration. Symmetry-adjusted weighting counteracts bias from individual models. For hyperparameter tuning, grid search was employed, while SHapley Additive exPlanations (SHAP) was used to quantify factor contributions. The performance of each model was evaluated using AUC and AP metrics. The key results show that all base models performed robustly (AUC > 0.8), with XGB showing high consistency (AUC = 0.927), and the performance of the symmetry-optimized ensemble (MLP + XGB) exceeded that of all the individual models (AUC = 0.964). The dominant drivers of Geohazards included elevation, slope, the topographic wetness index, and road adjacency, with high-susceptibility zones clustered in southeastern high-altitude terrain, central mountains, and road-intensive north-central sectors. The approach presented here provides an ensemble method based on predictive symmetry quantification that is effective under the constraints of small sample sizes. Full article

The widespread adoption of electric buses represents a major step forward in sustainable transportation, but also brings new operational challenges, particularly in terms of improving their efficiency and controlling costs. Therefore, battery energy consumption management is a key approach for addressing these issues. Accurate prediction of energy consumption and interpretation of the influencing factors are essential for improving operational efficiency, optimizing energy use, and reducing operating costs. Although existing studies have made progress in battery energy consumption prediction, challenges remain in achieving high-precision modeling and conducting a comprehensive analysis of the influencing features. To address these gaps, this study proposes a two-layer stacking framework for estimating the energy consumption of electric buses. The first layer integrates the strengths of three nonlinear regression models—RF (Random Forest), GBDT (Gradient Boosted Decision Trees), and CatBoost (Categorical Boosting)—to enhance the modeling capacity for complex feature relationships. The second layer employs a Linear Regression model as a meta-learner to aggregate the predictions from the base models and improve the overall predictive performance. The framework is trained on 2023 operational data from two electric bus routes (NO. 355 and NO. W188) in Changsha, China, incorporating battery system parameters, driving characteristics, and environmental variables as independent variables for model training and analysis. Comparative experiments with various ensemble models demonstrate that the proposed stacking framework exhibits superior performance in data fitting. Furthermore, XGBoost (Extreme Gradient Boosting, version 2.1.4) is introduced as a surrogate model to approximate the decision logic of the stacking framework, enabling SHAP (SHapley Additive exPlanations) analysis to quantify the contribution and marginal effects of influencing features. The proposed stacked and surrogate models achieved superior battery energy consumption prediction accuracy (lowest MSE, RMSE, and MAE), significantly outperforming benchmark models on real-world datasets. SHAP analysis quantified the overall contributions of feature categories (battery operation parameters: 56.5%; driving characteristics: 42.3%; environmental data: 1.2%), further revealing the specific contributions and nonlinear influence mechanisms of individual features. These quantitative findings offer specific guidance for optimizing battery system control and driving behavior. Full article

Banana de-handing is a critical yet labor-intensive step in postharvest processing, with current manual methods resulting in high costs and occupational risks. This study addresses the automation of de-handing point localization by integrating high-resolution 3D scanning and morphometric analysis of banana crowns with machine learning techniques. A total of 210 crown samples were analyzed to extract key morphological features, including inner arc length (L_i), inner arc radius (R_i), outer arc radius (R_o), and the distance between inner and outer arcs (D_oi), among others. Four machine learning algorithms, namely, Multi-Layer Perceptron (MLP), Gradient Boosted Decision Trees (GBDT), Extreme Gradient Boosting (XGBoost), and Random Forest (RF), were developed to predict the target radius (R_t) and target distance (D_ti) of the de-handing point. The RF models achieved the optimal predictive performance on the testing set, with the following results: for R_t, R² = 0.95, MAE = 1.50, and RMSE = 1.94; for D_ti, R² = 0.91, MAE = 1.33, and RMSE = 1.66. A Shapley Additive Explanations (SHAP) analysis revealed that L_i, R_i, and R_o were the most influential features for R_t, while D_oi was the most important for D_ti. Notably, feature threshold effects were observed, with limited gains in prediction accuracy beyond specific morphological values. These results provide a quantitative foundation for vision-guided automated de-handing systems, advancing intelligent and efficient banana postharvest management. Full article

To address the problems of traditional methods that rely on destructive sampling, the poor adaptability of fixed equipment, and the susceptibility of single-view angle measurements to occlusions, a non-destructive and portable device for three-dimensional phenotyping and biomass detection in lettuce was developed. Based on the Structure-from-Motion Multi-View Stereo (SFM-MVS) algorithms, a high-precision three-dimensional point cloud model was reconstructed from multi-view RGB image sequences, and 12 phenotypic parameters, such as plant height, crown width, were accurately extracted. Through regression analyses of plant height, crown width, and crown height, and the R² values were 0.98, 0.99, and 0.99, respectively, the RMSE values were 2.26 mm, 1.74 mm, and 1.69 mm, respectively. On this basis, four biomass prediction models were developed using Adaptive Boosting (AdaBoost), Support Vector Regression (SVR), Gradient Boosting Decision Tree (GBDT), and Random Forest Regression (RFR). The results indicated that the RFR model based on the projected convex hull area, point cloud convex hull surface area, and projected convex hull perimeter performed the best, with an R² of 0.90, an RMSE of 2.63 g, and an RMSEn of 9.53%, indicating that the RFR was able to accurately simulate lettuce biomass. This research achieves three-dimensional reconstruction and accurate biomass prediction of facility lettuce, and provides a portable and lightweight solution for facility crop growth detection. Full article

Background/Objectives: The resistance mutations EGFR^{L858R/T790M/C797S} in epidermal growth factor receptor (EGFR) are key factors in the reduced efficacy of Osimertinib. Predicting the inhibitory effects of Osimertinib derivatives against these mutations is crucial for the development of more effective inhibitors. This study aims to predict the inhibitory effects of Osimertinib derivatives against EGFR^{L858R/T790M/C797S} mutations. Methods: Six models were established using heuristic method (HM), random forest (RF), gene expression programming (GEP), gradient boosting decision tree (GBDT), polynomial kernel function support vector machine (SVM), and mixed kernel function SVM (MIX-SVM). The descriptors for these models were selected by the heuristic method or XGBoost. Comprehensive learning particle swarm optimizer was adopted to optimize hyperparameters. Additionally, the internal and external validation were performed by leave-one-out cross-validation ($Q_{LOO}^2$), 5-fold cross validation ($Q_{5-fold}^2$) and concordance correlation coefficient (CCC), $Q_{F1}^2$, and $Q_{F2}^2$. The properties of novel EGFR inhibitors were explored through molecular docking analysis. Results: The model established by MIX-SVM whose kernel function is a convex combination of three regular kernel functions is best: $R^2$ and RMSE for training set and test set are 0.9445, 0.1659 and 0.9490, 0.1814, respectively; $Q_{LOO}^2$, $Q_{5-fold}^2$, CCC, $Q_{F1}^2$, and $Q_{F2}^2$ are 0.9107, 0.8621, 0.9835, 0.9689, and 0.9680. Based on these results, the IC₅₀ values of 162 newly designed compounds were predicted using the HM model, and the top four candidates with the most favorable physicochemical properties were subsequently validated through PEA. Conclusions: The MIX-SVM method will provide useful guidance for the design and screening of novel EGFR^{L858R/T790M/C797S} inhibitors. Full article

Accurate estimation of forest carbon stock is critical for understanding ecosystem carbon dynamics and informing climate mitigation strategies. This study presents a deep learning framework that integrates Sentinel-2 multispectral imagery with an enhanced residual neural network for estimating aboveground forest carbon stock in the Liuchong River Basin, Bijie City, Guizhou Province, China. The proposed model incorporates multiscale residual blocks and channel attention mechanisms to improve spatial feature extraction and spectral dependency modeling. A dataset of 150 ground inventory plots was employed for supervised training and validation. Comparative experiments with Random Forest, Gradient Boosting Decision Trees (GBDT), and Vision Transformer (ViT) demonstrate that the enhanced ResNet achieves the best performance, with a root mean square error (RMSE) of 23.02 Mg/ha and a coefficient of determination ($R^2$) of 0.773 on the test set. Spatial mapping results further reveal that the model effectively captures fine-scale carbon stock variations across mountainous forested landscapes. These findings underscore the potential of combining multispectral remote sensing and advanced neural architectures for scalable, high-resolution forest carbon estimation in complex terrain. Full article

The micro-scale interplay between the built environment and innovation has attracted increasing scholarly attention. However, discussions on how such microdynamics operate and vary across high-density cities remain insufficient. This study focuses on nine high-density urban centres along the G60 S&T Innovation Valley and employs a fine-grained grid unit, viz. 1 km × 1 km, combined with the gradient boosting decision tree (GBDT) model to address these issues. Results show that urban construction density-related variables, including the building density, floor area ratio, and transportation network density, generally rank higher than the amenity density and proximity-related variables. The former contributes 50.90% of the total relative importance in predicting invention patent application density (IPAD), while the latter two contribute 13.64% and 35.46%, respectively. Threshold effect analysis identifies optimal levels for enhancing IPAD. Specifically, the optimal building density is approximately 20%, floor area ratio is 5, and transportation network density is 8 km/km². Optimal distances to universities, city centres, and transportation hubs are around 1 km, 17 km, and 9 km, respectively. Furthermore, significant city-level heterogeneity was observed: most density-related variables consistently have an overall positive association with IPAD, with metropolitan cities (e.g., Hangzhou and Suzhou) exhibiting notably higher optimal values compared to medium and small cities (e.g., Xuancheng and Huzhou). In contrast, the threshold effects of proximity-related variables on IPAD are more complex and diverse. These findings offer empirical support for enhancing innovation in high-density urban environments. Full article

Accurate forecasting of carbon emissions at the county level is critical to support China’s dual-carbon goals. However, most current studies are limited to national or provincial scales, employing traditional statistical methods inadequate for capturing complex nonlinear interactions and spatiotemporal dynamics at finer resolutions. To overcome these limitations, this study develops and validates a high-resolution predictive model using advanced gradient boosting algorithms—Gradient Boosting Decision Tree (GBDT), Extreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LightGBM)—based on socioeconomic, industrial, and environmental data from 2732 Chinese counties during 2008–2017. Key variables were selected through correlation analysis, missing values were interpolated using K-means clustering, and model parameters were systematically optimized via grid search and cross-validation. Among the algorithms tested, LightGBM achieved the best performance (R² = 0.992, RMSE = 0.297), demonstrating both robustness and efficiency. Spatial–temporal analyses revealed that while national emissions are slowing, the eastern region is approaching stabilization, whereas emissions in central and western regions are projected to continue rising through 2027. Furthermore, SHapley Additive exPlanations (SHAP) were applied to interpret the marginal and interaction effects of key variables. The results indicate that GDP, energy intensity, and nighttime lights exert the greatest influence on model predictions, while ecological indicators such as NDVI exhibit negative associations. SHAP dependence plots further reveal nonlinear relationships and regional heterogeneity among factors. The key innovation of this study lies in constructing a scalable and interpretable county-level carbon emissions model that integrates gradient boosting with SHAP-based variable attribution, overcoming limitations in spatial resolution and model transparency. Full article

Land use change plays a pivotal role in understanding surface processes and environmental dynamics, exerting considerable influence on regional ecosystem management. Traditional monitoring approaches, which often rely on manual sampling and single spectral features, exhibit limitations in efficiency and accuracy. This study proposes an innovative technical framework that integrates automated sample generation, multi-feature optimization, and classification model refinement to enhance the accuracy of land use classification and enable detailed spatiotemporal analysis in the Ebinur Lake Basin. By integrating Landsat data with multi-temporal European Space Agency (ESA) products, we acquired 14,000 pixels of 2021 land use samples, with multi-temporal spectral features enabling robust sample transfer to 12028 pixels in 2011 and 10,997 pixels in 2001. Multi-temporal composite data were reorganized and reconstructed to form annual and monthly feature spaces that integrate spectral bands, indices, terrain, and texture information. Feature selection based on the Gini coefficient and Out-Of-Bag Error (OOBE) reduced the original 48 features to 23. In addition, an object-oriented Gradient Boosting Decision Tree (GBDT) model was employed to perform accurate land use classification. A systematic evaluation confirmed the effectiveness of the proposed framework, achieving an overall accuracy of 93.17% and a Kappa coefficient of 92.03%, while significantly reducing noise in the classification maps. Based on land use classification results from three different periods, the spatial distribution and pattern changes of major land use types in the region over the past two decades were investigated through analyses of ellipses, centroid shifts, area changes, and transition matrices. This automated framework effectively enhances automation, offering technical support for accurate large-area land use classification. Full article

This study introduces an innovative dog emotion classification system that integrates four non-invasive physiological indicators—skin potential (SP), muscle potential (MP), respiration frequency (RF), and voice pattern (VP)—with the extreme gradient boosting (XGBoost) algorithm. A four-breed dataset was meticulously constructed by recording and labeling physiological signals from dogs exposed to four fundamental emotional states: happiness, sadness, fear, and anger. Comprehensive feature extraction (time-domain, frequency-domain, nonlinearity) was conducted for each signal modality, and inter-emotional variance was analyzed to establish discriminative patterns. Four machine learning algorithms—Neural Networks (NN), Support Vector Machines (SVM), Gradient Boosting Decision Trees (GBDT), and XGBoost—were trained and evaluated, with XGBoost achieving the highest classification accuracy of 90.54%. Notably, this is the first study to integrate a fusion of two complementary electrophysiological indicators—skin and muscle potentials—into a multi-modal dataset for canine emotion recognition. Further interpretability analysis using Shapley Additive exPlanations (SHAP) revealed skin potential and voice pattern features as the most contributive to model performance. The proposed system demonstrates high accuracy, efficiency, and portability, laying a robust groundwork for future advancements in cross-species affective computing and intelligent animal welfare technologies. Full article

Accurate prediction of the dynamic compressive strength of brittle engineering materials is of significant theoretical and engineering importance for underground engineering design, safety assessment, and dynamic hazard prevention. To enhance prediction accuracy and model interpretability, this study proposes a novel framework integrating stacking ensemble learning with SHapley Additive exPlanations (SHAP) for dynamic strength prediction. Leveraging multidimensional input variables, including static strength, strain rate, P-wave velocity, bulk density, and specimen geometry parameters, we constructed six machine learning regression models: K-Nearest Neighbors (KNN), Random Forest (RF), Gradient Boosting Decision Tree (GBDT), LightGBM, XGBoost, and Multilayer Perceptron Neural Network (MLPNN). Through comparative performance evaluation, optimal base models were selected for stacking ensemble training. Results demonstrate that the proposed stacking model outperforms individual models in prediction accuracy, stability, and generalization capability. Further SHAP-based interpretability analysis reveals that strain rate dominates the prediction outcomes, with its SHAP values exhibiting a characteristic nonlinear response trend. Additionally, structural and mechanical variables such as static strength, P-wave velocity, and bulk density demonstrate significant positive contributions to model outputs. This framework provides a robust tool for intelligent prediction and mechanistic interpretation of the dynamic strength of brittle materials. Full article

Microplastics are widely present in the environment, and their potential risks to human health have attracted increasing attention. Research on microplastics has exhibited exponential growth since 2014, with a fast-growing focus on human health risks. Keyword co-occurrence networks indicate a research shift from environmental pollution toward human exposure and health effects. Additionally, Trend Factor analysis reveals emerging research topics such as reproductive toxicity, neurotoxicity, and impacts on gut microbiota. This meta-analysis included 125 studies comprising 2977 data samples. The results demonstrated that cytotoxicity in experimental systems was primarily concentrated in Grade I (non-toxic, 62.8%) and Grade II (mildly toxic, 27.6%). Notably, inhibitory effects on cells were significantly enhanced when microplastic concentrations exceeded 40 μg/mL or particle sizes were smaller than 0.02 μm. The Gradient Boosting Decision Tree (GBDT) model was applied to predict cell viability, achieving an R² value of 0.737 for the test set and a classification accuracy of 81.5%. Furthermore, reproductive- and circulatory-system cells exhibited the highest sensitivity to microplastics, whereas connective-tissue cells had the lowest survival rates. The study also identified an overuse of polystyrene (PS) polymers and spherical particles in experimental designs, deviating from realistic exposure scenarios. Full article

Cadmium (Cd) pollution poses a severe threat to rice safety and human health, while traditional linear models exhibit significant limitations in predicting rice Cd accumulation due to environmental complexities. This study systematically evaluated the predictive performance of Random Forest (RF), Gradient Boosting Decision Tree (GBDT), and Residual Neural Networks (ResNet), using a multi-source soil–rice dataset comprising 57,200 samples from Hunan Province. The results showed that the RF model performed best on the test set (R² = 0.62), with the dominant features being soil’s available Cd (contributing 9.74%) and precipitation during the rice-filling stage (joint contribution of 15.96%). However, the model’s predictive performance experienced a sharp decline on the independent 2023 validation set comprising 393 samples from Yizhang County and Lengshuitan District, with R² values ranging from −0.12 to −0.31. This highlighted the fundamental limitations of static data-driven paradigms. Agronomic management measures, simplified by heterogeneous data and binary encoding, failed to effectively represent the actual intervention intensity. The study demonstrated that while machine learning models captured nonlinear relationships in laboratory environments, they struggled to adapt to the dynamic interactions and spatiotemporal heterogeneity of farmland systems. Future efforts should focus on developing hybrid models guided by mechanistic insights, integrating dynamic environmental processes and real-time data, and promoting localized “one model per region” strategies to enhance predictive robustness. This study provides methodological insights for the technological transformation of agricultural artificial intelligence, emphasizing that the deep integration of data-driven approaches and mechanistic understanding is crucial for overcoming the “last mile” challenge. Full article