Relative Importance Of Variables Research Articles

Deep learning-based quantitative analyses of feedback in the land-atmosphere interactions over the Vietnamese Mekong Delta

During the past several decades, the Vietnamese Mekong Delta (VMD) has experienced many severe droughts, resulting in significant impacts on both agriculture and aquaculture. In the evolution and intensification of droughts, local feedbacks in the Land-Atmosphere (LA) interactions were considered to play a crucial role. It is critical to quantify the impact of LA variables on drought processes and severity with the feedback loop of water and energy balances (e.g., soil moisture-latent and sensible heat-precipitation). In this study, a deep learning model, named Long- and Short-term Time-series Network (LSTNet), was applied to simulate the LA interactions over the VMD. With the ERA5 data as modelling inputs, the role of each key variable (e.g., soil moisture, sensible and latent heat) in the LA interactions over the period of 2011–2020 was quantified, and the variations of their inter-relationships in the future period (2015–2099) were also investigated based on the CMIP6 data. The LSTNet model has demonstrated that the deep learning algorithm can effectively capture the relative importance of key variables in the LA interactions. We found that it is crucial to evaluate the effects of soil moisture and sensible heat on the LA interactions, particularly in the dry periods when negative anomalies in soil moisture and sensible heat would significantly reduce the amount of precipitation. In addition, the decline in soil moisture and the rise in sensible heat are anticipated to further diminish precipitation in the future under the changing climate.

Machine learning-assisted performance prediction from the synthesis conditions of nanofiltration membranes

Optimizing membrane fabrication conditions, including monomer and catalyst concentrations, reaction time, etc., is crucial for achieving high-performance membranes. However, the multitude of variables involved can result in a complex and laborious permutation process, making trial-and-error optimization challenging. To address this issue, we developed a conventional multiple linear regression model and machine learning (ML) models, specifically random forest (RF) and multi-layer perceptron (MLP), to predict the performance of nanofiltration membranes in terms of key parameters such as pure water permeance or PWP (LMH/bar), Na2SO4 rejection (%), and NaCl rejection (%), based on specified input variable ranges. The input variables utilized to construct the models are the concentration of piperazine (PIP) (0.01–2 wt%), the concentration of trimesoyl chloride (TMC) (0.05–0.15 wt%), the concentration of sodium lauryl sulfate (SLS) (0–10 mM), and reaction time (5–1200 s). The conventional regression model failed to predict the membrane performance. The RF model exhibited the best prediction capability for PWP and NaCl rejection with the R2 of 0.9806 and 0.9812 for training, and 0.9669 and 0.9082 for testing, respectively. Similarly, the MLP model outperformed the RF model in predicting Na2SO4 rejection with an R2 of 0.9972 for testing and 0.9844 for training. The best membrane exhibited permeance ranging from 16 to 20 LMH/bar and selectivity between NaCl and Na2SO4 of over 4000, was facilitated by specific conditions. These conditions included lower concentrations of PIP (0.05–0.1 wt%), intermediate concentration of TMC (0.1 wt%), lower concentration of SLS (1 mM), and shorter reaction times (5–30 s). The best-performing ML-based models have been used to provide insights into the relative importance of these input variables in determining membrane performance, thereby aiding in correlating model predictions with fundamental principles of membrane fabrication processes.

Importance of hospital and clinical factors for early mortality in Takotsubo syndrome: Insights from the Swedish Coronary Angiography and Angioplasty Registry

BackgroundTakotsubo syndrome (TTS) is an acute heart failure syndrome with symptoms similar to acute myocardial infarction. TTS is often triggered by acute emotional or physical stress and is a significant cause of morbidity and mortality. Predictors of mortality in patients with TS are not well understood, and there is a need to identify high-risk patients and tailor treatment accordingly. This study aimed to assess the importance of various clinical factors in predicting 30-day mortality in TTS patients using a machine learning algorithm.MethodsWe analyzed data from the nationwide Swedish Coronary Angiography and Angioplasty Registry (SCAAR) for all patients with TTS in Sweden between 2015 and 2022. Gradient boosting was used to assess the relative importance of variables in predicting 30-day mortality in TTS patients.ResultsOf 3,180 patients hospitalized with TTS, 76.0% were women. The median age was 71.0 years (interquartile range 62–77). The crude all-cause mortality rate was 3.2% at 30 days. Machine learning algorithms by gradient boosting identified treating hospitals as the most important predictor of 30-day mortality. This factor was followed in significance by the clinical indication for angiography, creatinine level, Killip class, and age. Other less important factors included weight, height, and certain medical conditions such as hyperlipidemia and smoking status.ConclusionsUsing machine learning with gradient boosting, we analyzed all Swedish patients diagnosed with TTS over seven years and found that the treating hospital was the most significant predictor of 30-day mortality.

Disentangling ecological drivers of interspecific achromatic plumage variation in birds

AbstractAimUnderstanding the ecological determinants of interspecific achromatic (light‐to‐dark) plumage variation in birds is crucial yet challenging due to the complex interplay of climatic, habitat‐related, and morphological influences. This study aimed to disentangle the effects of temperature, precipitation, habitat openness, body mass and hand‐wing index (HWI, a widely used single‐parameter proxy for the extent to which a species relies on flight) on shaping achromatic plumage variation among bird species.LocationGlobal.Time PeriodContemporary.Major Taxa StudiedBirds.MethodsBased on data from over 8000 sessile bird species globally, we employed phylogenetic linear regressions to account for achromatic plumage colour in relation to temperature, precipitation, habitat openness, body mass and HWI, while correcting for phylogenetic non‐independence between species. Furthermore, we conducted phylogenetic path analyses to decompose direct from indirect effects.ResultsWe found that temperature, precipitation, habitat openness and body mass exerted separate but interactive effects on the variation in achromatic colour across species. Species inhabiting cold, wet or densely vegetated environments were darker coloured, while smaller species were lighter. Darker plumage was more strongly related to higher precipitation in colder regions for nocturnal species. For diurnal species, darker plumage was more closely associated with higher precipitation in more open habitats, whereas lighter plumage was more linked to lower mass in denser habitats. Noteworthy was the identification of a substantial correlation between achromatic colour and HWI. Diurnal species that are more aerial were lighter. Conversely, nocturnal flyers, especially females, tended to be darker.Main ConclusionsThe findings highlight the multifaceted nature of plumage coloration evolution, with adaptations for thermal efficiency, crypsis, signalling, waterproofing or protection against bacteria. However, the variable relative importance of these factors among groups emphasizes the significance of each factor in different contexts.

Brand perception and its relationships to satisfaction with a virtual sporting event

PurposeThe purpose of this study is to use structural equation modeling to examine the interaction between the variables of corporate image, credibility, trust and satisfaction in the context of a virtual sporting event. The aim is to determine whether these variables have significant relationships with each other and which of them has the greatest influence on the prediction of participants' satisfaction.Design/methodology/approachA structured questionnaire was used, based on previously validated scales. The survey was administered using the LimeSurvey platform. The sample consisted of a total of 588 participants of the Medellín virtual marathon.FindingsThe results of the study reveal significant findings regarding the relationships between the variables of corporate image, credibility, trust, and satisfaction in virtual sporting events. In particular, it is highlighted that trust emerges as the most influential factor in participants' satisfaction, which offers an insightful understanding of the importance of this variable in the user experience in virtual sporting events.Research limitations/implicationsThis study emphasizes the importance of brand analysis in the sports environment, stressing that the actions undertaken by managers should highlight both the corporate image and the connections with users, given their fundamental role in customer satisfaction. Likewise, the study of these variables within the sports context provides new knowledge and fills existing gaps within the academy. Limitations include the sample and the lack of consideration of all brand variables.Practical implicationsThe need to cultivate a strong and well-managed image to build trust with participants is emphasized for organizers of virtual sporting events. It is crucial to work on establishing long-term credibility, especially in the relatively new context of virtual racing. Maintaining, and building the virtual career offering is essential to strengthening relationships, demonstrating a robust corporate image. In addition, since trust and credibility have a significant impact on participant satisfaction in this type of event, managers must communicate the assurance that virtual careers offer an experience free of uncertainty and risk, which is particularly attractive to a new customer base interested in this format.Originality/valueThis article presents an original contribution by investigating the relationships between corporate image, credibility, trust, and satisfaction in the context of virtual sporting events. It employs a structural equation model to assess the significance and predictive capacity of these variables. Notably, the study identifies trust as the most influential factor in predicting participant satisfaction. These findings offer valuable insights into the relative importance of brand variables in shaping user satisfaction within the virtual sporting event domain. By shedding light on these dynamics, the research aids event managers in making informed resource allocation decisions, contributing to a nuanced understanding of brand impact in this context.

Fire resistance time prediction and optimization of cold-formed steel walls based on machine learning

Many full-scale experiments and numerical studies have been conducted to determine the fire performance of cold-formed steel (CFS) walls, but these studies are expensive and time consuming. This study proposes a machine learning (ML) based framework aiming at accurately predicting the fire resistance time (FRT) and optimizing the design of CFS walls under ISO 834 fire condition. To overcome the limitation of 32 experimental data points in the literature, a validated numerical method was used to generate 592 data points to expand the dataset of CFS walls, considering different wall configurations, various sheathing board types and thicknesses. The XGBoost (eXtreme Gradient Boosting) model was trained with numerical data and tested with experimental data. The hyperparameter tuning of the XGBoost model was implemented with Bayesian optimization, and it was found that the XGBoost model accurately predicted the FRT of CFS walls, with R2 and MAPE values being 0.933 and 8.46 %, respectively. The prediction process of the XGBoost model was interpreted by the SHAP (SHapley Additive exPlanations) method to determine the relative importance of input variables. The NSGA-II algorithm was adopted to optimize the FRT-cost dual-objective of CFS walls and the Pareto front including optimal solutions was emerging. The cost for each solution of Pareto front sets was lower than that in the real dataset at same FRT level. This phenomenon implied that the ML-based optimization framework successfully identified the cost-efficient design process. The proposed ML-based optimization framework offers a promising alternative for engineers to design CFS walls effectively with both mechanical and economic objectives.

Utility of Body Weight, Urine Color, and Thirst Perception (WUT) in Determining Hydration in Young Adults.

The primary aim of this study was to assess the efficacy of the weight, urine, thirst (WUT) framework in predicting dehydration after a body water manipulation protocol, while concurrently determining the individual and interactive contributions of the model components. The total study sample was 93 participants (female, n = 47), recruited from two institutions. Phase 1 involved collecting daily hydration measures from free-living participants (Study 1, 58 participants for 3 days; Study 2, 35 participants for 7 days). Phase 2 entailed a two-hour passive heating protocol, where participants from Study 2 were randomly assigned to one of three groups that manipulated total body water over 24-hours using passive heating and fluid restriction. During each Phase, participants provided urine samples, underwent body mass measurements, and completed questionnaires pertaining to thirst perception. Morning and 24-hour urine samples were assessed for color, osmolality, and specific gravity. Differences between intervention groups, based on the probability of hydration status, were examined (ANOVA) and ridge regression analysis assessed the relative importance of variables within the WUT model. The study revealed significant differences among the intervention groups for predicted probability of dehydration, as determined by changes in body mass (p = 0.001), urine color (p = 0.044), and thirst perception (p < 0.001). Binomial ridge regression indicated that change in body mass (58%) and thirst perception (26%) were the most influential predictors of dehydration. These data support use of an enhanced version of the WUT model, underscoring the significance of changes in body mass and thirst perception in the assessment of hydration status.

Old-growth forests are critical to safeguard tropical birds in complex landscape mosaics exposed to slash-and-burn agriculture

ContextSlash-and-burn agriculture generates landscape mosaics composed of different land uses. Ecological theory postulates that in these mosaics, the structure of species assemblages depends on both local- and landscape-scale factors, but their relative role remains poorly known.ObjectivesUnderstanding the relative importance of local and landscape variables for bird assemblages in landscape mosaics of the Mayan tropical forest, Mexico.MethodsWe evaluated the effect of two local variables (land use type and tree basal area) and two landscape variables (old-growth forest cover and human proximity) on bird abundance and diversity, separately assessing forest and non-forest birds. As species extirpation may accelerate when forest loss exceeds certain limits (extinction thresholds), we assessed forest loss effects with linear and nonlinear models.ResultsWe recorded 1867 birds from 118 species in four land-use types (old-growth forest, secondary forest, burned lands, and agricultural lands). Land-use type and surrounding old-growth forest cover better predicted bird abundance and diversity than tree basal area and human proximity. Agricultural lands showed the lowest abundance and diversity of forest species, but the highest abundance and diversity of non-forest species. We found no support for the extinction threshold hypothesis. Yet, independently of the land use, old-growth forest loss decreased the diversity of forest species and increased the abundance and diversity of non-forest species. Human proximity also decreased total bird abundance, especially because of the loss of forest species.ConclusionsOur findings highlight the key role played by old-growth forests for preserving bird assemblages in landscape mosaics exposed to slash-and-burn agriculture.

Probabilistic UK Climate Projections Conditioned on Global Warming Levels

AbstractProbabilistic projections from the UK Climate Projections 2018 are presented for four global warming levels (GWLs) at 1.5, 2, 3, and 4°C above the 1850–1900 baseline. Our results show how uncertainties associated with climate models and four representative concentration pathways (RCP) emission scenarios translate to UK regional scale changes in maximum temperature and precipitation, with data also available for minimum and mean temperatures, humidity and surface net downward shortwave radiation flux. We compare weighting the likelihood of RCPs based on (hypothetical) policy decisions, against our baseline assumption that each RCP is equally likely. Differences between weighted and unweighted GWL distributions are small, particularly in relation to the full breadth of uncertainties that are incorporated into the probabilistic projections. Finally we quantify the relative importance of scenario, model and internal variability on regional projected GWLs and show that uncertainty associated with an uncertain climate response to forcings dominates at all GWLs.

Enlarged perivascular spaces are associated with brain microangiopathy and aging in multiple sclerosis

Background: Growing evidence links brain-MRI enlarged perivascular spaces (EPVS) and multiple sclerosis (MS), but their role remains unclear. Objective: This study aimed to investigate the cross-sectional associations of EPVS with several neuroinflammatory and neurodegenerative features in a large multicentric-MS cohort. Methods: In total, 207 patients underwent 3T axial-T2-weighted brain-MRI for EPVS assessment (EPVS dichotomized into high/low according to ⩾ 2/< 2 rating categories). MRI biomarkers included brain-predicted age and brain-predicted age difference (brain-PAD), central vein sign (CVS)-positive lesion percentage (CVS%), paramagnetic rim and cortical lesions, T2-lesion load, and brain volumetry. The variable relative importance for EPVS-category prediction was explored using a classification random forest approach. Results: High EPVS patients were older (49 vs 44 years, p = 0.003), had ⩾ 1 vascular risk factors (VRFs; p = 0.005), lower CVS% (67% vs 78%, p < 0.001), reduced brain volumes (whole brain: 0.63 vs 0.73, p = 0.01; gray matter: 0.36 vs 0.40; p = 0.002), and older brain-predicted age (58 vs 50 years, p < 0.001). No differences were found for neuroinflammatory markers. After adjusting for age and VFRs (multivariate analyses), the high EPVS category correlated with lower CVS% (odds ratio (OR) = 0.98, 95% confidence interval (CI) = 0.96–0.99; p = 0.02), lower whole brain (OR = 0.01, 95% CI = 0.0003–0.5; p = 0.02), gray matter (OR = 0.0004, 95% CI = 0.0000004–0.4; p = 0.03) volumes, and higher brain-PAD (OR = 1.05, 95% CI = 1.01–1.09; p = 0.02). Random forest identified brain-PAD as the most important predictor of high EPVS. Conclusion: EPVS in MS likely reflect microangiopathic disease rather than neuroinflammation, potentially contributing to accelerated neurodegeneration.

How do short-term and long-term factors impact the aboveground biomass of grassland in Northern China?

The aboveground biomass (AGB) of grassland, a crucial indicator of productivity, is anticipated to widespread changes in key ecosystem attributes, functions and dynamics. Variations in grassland AGB have been extensively documented across various spatial and temporal scales. However, a precise method to disentangle long-term effects from short-term effects on grassland AGB and assess the attribution of explanatory factors for AGB change remains elusive. This study aimed to quantify the impact of key climatic factors, soil properties, and grazing intensity on grassland AGB changes, utilizing data spanning the 1980s and the 2000s in Northern China. The Co-regression model was explored to separate the long-term effects and short-term effects of grassland AGB, while the Generalized Linear Model (GLM) was utilized to analyze the contributions of key variables to AGB. This approach effectively avoids issues related to regression to the mean and mathematical coupling. The results revealed that the influence of climatic variables, soil texture and grazing intensity on grassland AGB changes could be decomposed into long-term, short-term and random effects. Long-term effects explained 73.6% of AGB variation, whereas short-term effect only accounted for 5.9% of AGB change. Additionally, the short-term effect was divided into direct and indirect effects, with the direct effect explaining 1.3% of AGB variation, and the indirect effect explained 4.6% of AGB dynamics. The relative importance of key variables in grassland AGB was assessed, identifying soil parameters and precipitation as the main driving factors in the study area. This study introduces a robust methodology to enhance model performance in distinguishing long-term and short-term effects on grassland AGB, contributing to the sustainable development of grassland ecology in similar regions.

Impact of surgeon variability on outcomes after total shoulder arthroplasty: an analysis of 2188 surgeons

BackgroundThere has been limited evaluation of surgeon-specific factors on outcomes after total shoulder arthroplasty (TSA). The goals of this study were (1) to determine the impact of surgeon procedural volume, career duration, practice type, and fellowship training on TSA outcomes at 2-year follow-up and (2) to evaluate the relative importance of surgeon variables in relation to patient variables in influencing outcomes. MethodsThe PearlDiver Mariner database was queried to identify all surgeons who performed a minimum of 11 TSA procedures from 2010 to 2018. An Internet search of publicly available data was performed to determine the career duration, practice type, and fellowship training of each surgeon. Multivariate logistic regression models were built to determine the relationship between surgeon-specific variables and 2-year surgical complications and revisions and 90-day readmissions. Variable importance of patient-specific and surgeon-specific factors was determined by the Akaike information criterion increase of these models. ResultsA total of 2188 surgeons who performed 93,122 TSA procedures were identified in this database. Higher reverse TSA surgical volume was associated with fewer surgical complications and revisions, although such a relationship was not observed for anatomic TSA. Revision after anatomic TSA was more common among surgeons who were in their first 5 years of practice. There was a higher risk of surgical complication and readmission among academic surgeons, but this did not equate with a higher risk of revision. The impact of these surgeon-specific factors on outcomes was small in relation to patient-level variables, such as age, sex, and number of medical comorbidities. ConclusionSurgeon procedural volume, career duration, and practice setting influence the surgical complication, revision, and readmission rates after TSA. The impact of surgeon factors was small relative to patient variables, highlighting the importance of patient selection in mitigating adverse outcomes.

The snow cover is more important than other climatic variables on the prediction of vegetation dynamics in the Pyrenees (1981–2014)

The dynamics of the mountain vegetation is governed by multiple climatic drivers including temperature, precipitation, radiation and snow cover variability. However, in the Mediterranean environment, little is known about the relative importance of each variable. In this study we assess how different snowpack indices (the maximum annual accumulation, the length of the snow season, and the melt-out date) and key climate variables (precipitation, temperature and shortwave solar radiation) control the interannual variability of the maximum Normalized Difference Vegetation Index (peak NDVI) in the Pyrenees. We use a 33 year long remote sensing dataset (1981–2014) to build a statistical model relating the annual peak NDVI with snow and climate variables. In elevated areas characterized by a well developed seasonal snowpack the melt-out date was the most important climatic variable for predicting the annual peak NDVI. However, at lower elevations where snow presence is ephemeral, shortwave solar radiation was the most important variable. This change in the relative importance of climatic variables occurs around 1300 m a.s.l. The results do not show a significant contribution of maximum snow accumulation, suggesting that indicators of snow presence (i.e. melt-out date or snow season duration), which are significantly easier to obtain than snow mass indicators from remote sensing, could be used to model the influence of the snowpack on peak NDVI at regional scale.

Evaluating spray drift from Uncrewed Aerial Spray Systems: A machine learning and variance-based sensitivity analysis of environmental and spray system parameters

Uncrewed Aerial Spray Systems (UASS), commonly called drones, have become an important application technique for plant protection products in Asia and worldwide. As such, environmental variables and spray system parameters influencing spray drift deserve detailed investigations. This study presents the data analysis of 114 UASS drift trials conducted between December 2021 and December 2022 in China. Study design was based on the ISO 22866:2005 protocol for spray drift trials and considered different UASS platforms, nozzles, and release heights, and specifically continuously measured weather conditions. The relative importance of the environmental variables and spray system parameters was evaluated by a random forest (RF) feature importance analysis, a Sobol sensitivity analysis and partial dependence plots. This approach was preferred to linear ranking techniques such as ANOVA (analysis of variance) due to the non-linearity of the system. In addition, partial dependence plots are proposed to visualize the relationship between specific input parameters within the system.Drift deposition curves calculated from the 114 trials show good agreement with previous UASS trials reported in the literature. As reported in previous studies, spray drift following UASS applications is lower than for manned aerial vehicles, greater than for ground spray applications, and similar to drift observed from orchard air blast applications. In addition, 9 trials were conducted on corn fields in order to evaluate the potential effect of crop cover on spray drift. Spray drift was observed to be reduced over the cropped soil, suggesting that plant cover might possibly reduce spray drift. These findings could help supporting drift mitigation policies, stewardship advice and product labelling around the world.

Taxonomic distinctness and diversity patterns of a polychaete (Annelida) community on the continental shelf of the Southern Gulf of Mexico.

The spatial patterns of taxonomic diversity of annelid polychaete species from the continental shelf in the Southern Gulf of Mexico were examined in this study. We used taxonomic distinctness and its spatial variations to explore the diversity patterns and how they change between Southern Gulf of Mexico regions. In addition, using taxonomic distinctness as a dissimilarity measure and Ward's Clustering, we characterized three distinct faunal assemblages. We also investigated patterns of richness, taxonomic distinctness, and distance decay of similarity between sampling stations as a ß-diversity measure. Finally, we examined the spatial relationships between polychaete assemblages and environmental variables to test the relative importance of spatial and environmental components in annelid polychaete community structure from the Southern Gulf of Mexico. We used a combination of eigenvector-based multivariate analyses (dbMEMs) and distance-based redundancy analysis (dbRDA) to quantify the relative importance of these explanatory variables on the spatial variations of taxonomic distinctness. The significance level of spatial and environmental components to the distribution of polychaete species showed that the combined effect of spatial processes and sediment characteristics explained a higher percentage of the variance than those parameters could alone.

Cluster-based regression using variational inference and applications in financial forecasting

This paper describes an approach to simultaneously identify clusters and estimate cluster-specific regression parameters from the given data. Such an approach can be useful in learning the relationship between input and output when the regression parameters for estimating output are different in different regions of the input space. Variational Inference (VI), a machine learning approach to obtain posterior probability densities using optimization techniques, is used to identify clusters of explanatory variables and regression parameters for each cluster. From these results, one can obtain both the expected value and the full distribution of predicted output. Other advantages of the proposed approach include the elegant theoretical solution and clear interpretability of results. The proposed approach is well-suited for financial forecasting where markets have different regimes (or clusters) with different patterns and correlations of market changes in each regime. In financial applications, knowledge about such clusters can provide useful insights about portfolio performance and identify the relative importance of variables in different market regimes. An illustrative example of predicting one-day S&P change is considered to illustrate the approach and compare the performance of the proposed approach with standard regression without clusters. Due to the broad applicability of the problem, its elegant theoretical solution, and the computational efficiency of the proposed algorithm, the approach may be useful in a number of areas extending beyond the financial domain.

Using machine learning models to estimate Escherichia coli concentration in an irrigation pond from water quality and drone-based RGB imagery data

The rapid and efficient quantification of Escherichia coli concentrations is crucial for monitoring water quality. Remote sensing techniques and machine learning algorithms have been used to detect E. coli in water and estimate its concentrations. The application of these approaches, however, is challenged by limited sample availability and unbalanced water quality datasets. In this study, we estimated the E. coli concentration in an irrigation pond in Maryland, USA, during the summer season using demosaiced natural color (red, green, and blue: RGB) imagery in the visible and infrared spectral ranges, and a set of 14 water quality parameters. We did this by deploying four machine learning models – Random Forest (RF), Gradient Boosting Machine (GBM), Extreme Gradient Boosting (XGB), and K-nearest Neighbor (KNN) – under three data utilization scenarios: water quality parameters only, combined water quality and small unmanned aircraft system (sUAS)-based RGB data, and RGB data only. To select the training and test datasets, we applied two data-splitting methods: ordinary and quantile data splitting. These methods provided a constant splitting ratio in each decile of the E. coli concentration distribution. Quantile data splitting resulted in better model performance metrics and smaller differences between the metrics for both the training and testing datasets. When trained with quantile data splitting after hyperparameter optimization, models RF, GBM, and XGB had R2 values above 0.847 for the training dataset and above 0.689 for the test dataset. The combination of water quality and RGB imagery data resulted in a higher R2 value (>0.896) for the test dataset. Shapley additive explanations (SHAP) of the relative importance of variables revealed that the visible blue spectrum intensity and water temperature were the most influential parameters in the RF model. Demosaiced RGB imagery served as a useful predictor of E. coli concentration in the studied irrigation pond

Elevational Patterns of Forest Evapotranspiration and Its Sensitivity to Climatic Variation in Dryland Mountains

Elevational climatic heterogeneity, complex terrains, and varying subsurface properties affect the sensitivity of evapotranspiration (ET) in dryland mountain forests to hydrometeorological changes. However, the elevational distribution of ET sensitivity and its major influencing factors remain poorly understood. This study focused on the mid-altitude zone (1000–3500 m) forests in the Chinese Western Tianshan Mountains and assessed ET sensitivity to multiple climate variables, including precipitation (P) and potential evapotranspiration (PET), from 2000 to 2020. To evaluate the multi-year mean and trends in ET sensitivity, multi-source remote sensing data and regional survey data were analyzed using Spearman’s correlation coefficient, the sliding window method, and Kendall’s test. Furthermore, the relative importance of environmental variables (topography, geology, soil, and vegetation) was investigated. P and PET showed no significant trends, while ET exhibited a significant increasing trend (5.81 mm/yr, p &lt; 0.01), particularly at elevations above 2000 m. Most forests (93.5%) showed a positive sensitivity of ET to P, and 70.0% showed a positive sensitivity of ET to PET, mainly at elevations of 1500–2500 m. Additionally, the trend in ET sensitivity to P decreased with an increasing elevation, with 64.5% showing a positive trend. Meanwhile, the trend in ET sensitivity to PET increased with elevation, with 88.1% showing a positive trend. Notably, 53.2% of the forests showed increasing ET sensitivity trends to both P and PET, primarily at elevations of 2000–3000 m with a mean normalized difference vegetation index (NDVI) of 0.56. Geological factors, particularly the hydrological properties of weathered bedrock, contributed the most (~47%) to mean sensitivity. However, geological and vegetative factors, including the NDVI and root zone water availability, were the main contributors (35% each) to the sensitivity. This study highlights the elevation-dependent sensitivity of dryland mountain forests to hydrothermal changes, with higher-elevation forests (&gt;2000 m) being more sensitive to global warming.

Machine learning strategy for light lamb carcass classification using meat biomarkers

In Mediterranean areas, lamb meat is considered to be of great commercial value. Moreover, consumers are becoming increasingly interested in understanding the origin of lamb meat and its associated production and breeding systems. Among many applications, algorithms based on artificial intelligence are used to identify the origin of food products, and in this context, algorithms such as the Support Vector Machine (SVM), K-Nearest Neighbours (KNN), and the Artificial Neural Network (ANN) have been proposed to differentiate the origin of the animals according to their feeding diet. The objective of this study was to evaluate the performance of a variable reduction method based on a multiple regression model and three widely-used machine learning algorithms (SVM, KNN and ANN) for the classification of three commercial light lamb carcasses, from three feeding diets, in an indigenous Spanish breed (Mallorquina), using fatty acid and volatile compound biomarkers of meat. Machine learning algorithms were employed to discriminate lamb carcasses using 14 identified significant biomarkers, which were arranged based on an estimation of the relative importance (stepwise forward multiple regression F-score) of the input variables. We achieved high performances for the SVM, KNN and ANN algorithms, with 86%, 98% and 98% prediction accuracy, respectively. Among the 14 biomarkers used, 7 were identified as showing the highest discriminant capacity. The F-scores indicate that C17:1 and C20:5 n-3 fatty acids, and 2,5-dimethylpyrazine and 3-methylbutanal volatile compounds are the four most relevant biomarkers for predicting three lamb feeding diets.

Machine Learning-Based Causality Analysis of Human Resource Practices on Firm Performance

An organization’s human resource management practices are essential for its competitive advantage. This study specifically examined human resource (HR) practices that predict corporate performance (employee turnover and firm sales) based on a backpropagation neural network (BPN)-based causality analysis. This study aims to test how to optimize human resource practices to improve organizational performance. This study elucidated the effect of HR practices and organizational-level factors on predicting employee turnover and firm sales. The BPN-based causality analysis revealed the relative importance of explanatory variables on firm performance. To test the model, it employed the Human Capital Corporate Panel open data on Korean companies’ HR practices and other characteristics. The analysis identifies causal relationships between specific HR practices and firm performance. The results show that compensation-related HR practices are most influential in predicting firm sales and employee turnover. Moreover, training-related HR practices were modest, and talent acquisition and performance management practices had relatively weak effects on the two outcomes. The study provides insights into how human resource practices can be optimized to improve firm performance and enhance organizational effectiveness. The findings of this study contribute to the growing body of research on the use of machine learning in HR management and suggest practical implications for managers’ insights to optimize HR practices.