Search

Traditionally, the design of an expert system involves acquiring knowledge, in the form of symbolic rules, directly from the expert(s), which is a complex and time‐consuming task. Although expert systems approach is quite old, it is still present, especially where explicit knowledge representation and reasoning, which assure interpretability and explainability, are necessary. Therefore, machine learning methods have been devised to extract rules from data, to facilitate that task. However, those methods are quite inflexible in adapting to the application domain and provide no help in designing the expert system. In this work, we present a framework and corresponding tool, namely ACRES, for semi‐automatically generating expert systems from datasets. ACRES allows for data preprocessing, which helps in structuring knowledge in the form of a tree, called rule hierarchy, which represents (possible) dependencies among data variables and is used for rule formation. This improves interpretability and explainability of the produced systems. We have also designed and evaluated alternative methods for rule extraction from data and for calculation and use of certainty factors, to represent uncertainty; CFs can be dynamically updated. Experimental results on seven well‐known datasets show that the proposed rule extraction methods are comparable to other popular machine learning approaches like decision trees, CART, JRip, PART, Random Forest, and so on, for the classification task. Finally, we give insights on two applications of ACRES. [ABSTRACT FROM AUTHOR]

This study develops an intelligent optimization method of the facility environment (i.e., road facilities and surrounding landscapes) from drivers' visual perception to adjust operation speeds on rural roads. Different from previous methods that heavily rely on expert experience and are time‐consuming, this method can rapidly generate optimized visual images of the facility environment and promptly verify the optimization effects. In this study, a visual road schema model is established to quantify the facility environment from drivers' visual perception, and an automated optimization scheme determination approach considering the original facility environment characteristics is proposed using self‐explaining theory. Then, Cycle‐consistent generative adversarial network is used to automatically generate optimized facility environment images. To verify the optimization effect, operation speeds of the optimized facility environments are predicted using random forest. The case study shows that this method can effectively optimize the facility environment where original operation speeds are more than 20% over the speed limits, and the whole process only takes 1 h far less than several months or years in previous ways. Overall, this study advances the intelligence level in optimizing the facility environment and enhances rural road safety. [ABSTRACT FROM AUTHOR]

Accurate and reliable forecasting of cancellations is important for successful revenue management in the tourism industry. The objective of this study is to develop classification models to predict hotel booking cancellations. The work involves a number of key steps, such as data preprocessing to properly prepare the data; feature engineering to identify relevant attributes to help improve the predictive ability of the models; hyperparameter settings of the models, including choice of optimizers and incorporation of dropout layers to avoid overfitting in the neural networks; potential overfitting is evaluated using K‐fold cross‐validation; and performance is analysed using the confusion matrix and various performance metrics. The algorithms used are Multilayer Perceptron Neural Network, Radial Basis Function Neural Network, Deep Neural Network, Decision Tree Classifier, Random Forest Classifier, Ada Boost Classifier and XgBoost Classifier. Finally, the results of all models are compared, visualizing Deep Neural Network and XgBoost as the most suitable models for predicting hotel reservation cancellations. [ABSTRACT FROM AUTHOR]

Aims: Although there are various model‐based approaches to individualized vancomycin (VCM) administration, few have been reported for adult patients with periprosthetic joint infection (PJI). This work attempted to develop a machine learning (ML)‐based model for predicting VCM trough concentration in adult PJI patients. Methods: The dataset of 287 VCM trough concentrations from 130 adult PJI patients was split into a training set (229) and a testing set (58) at a ratio of 8:2, and an independent external 32 concentrations were collected as a validation set. A total of 13 covariates and the target variable (VCM trough concentration) were included in the dataset. A covariate model was respectively constructed by support vector regression, random forest regression and gradient boosted regression trees and interpreted by SHapley Additive exPlanation (SHAP). Results: The SHAP plots visualized the weight of the covariates in the models, with estimated glomerular filtration rate and VCM daily dose as the 2 most important factors, which were adopted for the model construction. Random forest regression was the optimal ML algorithm with a relative accuracy of 82.8% and absolute accuracy of 67.2% (R2 =.61, mean absolute error = 2.4, mean square error = 10.1), and its prediction performance was verified in the validation set. Conclusion: The proposed ML‐based model can satisfactorily predict the VCM trough concentration in adult PJI patients. Its construction can be facilitated with only 2 clinical parameters (estimated glomerular filtration rate and VCM daily dose), and prediction accuracy can be rationalized by SHAP values, which highlights a profound practical value for clinical dosing guidance and timely treatment. [ABSTRACT FROM AUTHOR]

ABSTRACT Aim Location Time Period Major Taxa Studied Methods Results Main Conclusions The relative importance of traits and phylogeny to predict species extinction risk is unclear and it depends on which traits are measured and their phylogenetic conservatism. Here, we evaluate the power of functional traits, ecological characteristics, such as range size and specialization, and phylogeny to predict climatic risks in European butterflies.Europe.Distribution data from 1981 to 2002 was used to project range shifts up to 2080.In total, 268 European butterfly species (Rhopalocera).We extracted information on climatic risk for each species from the literature. Two hypothetical conditions were assumed: full climate tracking and no climate tracking. We used variation partitioning to quantify and disentangle the effects of phylogeny, four traits (wingspan, voltinism, overwintering and egg volume) and two ecological characteristics (range size and specialization) on climatic risk. We used Random Forest as an imputation method to predict climatic risk values for additional European butterfly species that have not been previously modelled.Range size and degree of specialization strongly predicted climatic risk of European butterflies, but only for no climate tracking scenarios. Overall, more generalist species have larger ranges and are less vulnerable to climate change. Furthermore, we found that phylogenetic information adds strong power to explain climatic risk, especially, but not exclusively for scenarios that assume full climate tracking.Considering current scenarios of habitat fragmentation in Europe, it is likely that species with smaller ranges and high degree of specialization will be unable to track their climates and thus be more at risk. However, our results imply that assisting dispersal and colonization might allow small ranged species to cope with climate change. Our analysis also indicate that phylogeny can be used as a proxy of unmeasured traits to predict climatic risk under certain circumstances. [ABSTRACT FROM AUTHOR]

Synthetic aperture radar (SAR) image change detection suffers from poor quality of the difference image and low detection accuracy. Hence, this paper proposes a SAR image change detection method based on a fused difference image and an optimized random forest scheme, termed LRN‐SSARF. Specifically, a fusion operator difference image LRN is proposed, which is generated using a weighted fusion of log‐ratio (LR), ratio (R), and normalized ratio (NoR). This difference image generation method reduces noise's influence. Then, the Otsu algorithm is applied to segment the difference image and select the training samples. The training samples are input into the random forest (RF) model optimised by the sparrow search algorithm (SSA) for training and classification. Finally, the region link is uesd to refine the detection results and generate the final result. The change detection results of six real SAR image scenes highlight that the proposed algorithm has a high detection accuracy, and affords appealing integrity and detailed information about the change regions. Specially, the detection accuracy advantage of the Bangladesh dataset is larger, with the accuracy and Kappa coefficient reaching 98.04% and 92.00%, much higher than the competitor methods. [ABSTRACT FROM AUTHOR]

As lake and reservoir ecosystems transition across major environmental regimes (e.g., mixing regime) resulting from anthropogenic change, setting predictive expectations is imperative. We tested the hypothesis that (dissolved) oxygen is more predictable in monomictic reservoirs that thermally stratify throughout the summer (warm) season compared to polymictic reservoirs that stratify intermittently. Using two‐hourly vertical profiles of oxygen, we compared daily‐aggregated errors of oxygen predictions from random forests across and within two monomictic and two polymictic reservoirs in the south‐central (subtropical) USA. Although one monomictic reservoir was typically more predictable than the polymictic reservoirs, the hypereutrophic, small monomictic reservoir had less predictable oxygen patterns potentially related to rapid oxygen cycling and intrusions of oxygenated waters in the hypolimnion without mixing. Daily mixing did not relate strongly to model errors. Water temperature, depth, and wind were the most important predictors, but were not clearly related to season or mixing. Lastly, we compared multiple model types (regression, neural network, and process‐based) in one polymictic reservoir to test how our interpretations of oxygen predictability were sensitive to model type, finding that the models generally agreed; however, the process‐based model poorly predicted oxygen in the middle of the vertical profiles (5 m) where most models performed poorly due to a temporally unstable, vacillating metalimnion. Our results suggest predicting reservoir oxygen dynamics may be easier in stratified reservoirs, but eutrophication and complex hydrodynamics may cause forecasting surprises especially for those who use or manage water resources in mono‐ or dimictic reservoirs. [ABSTRACT FROM AUTHOR]

Aim: Seamounts are conspicuous geological features with an important ecological role and can be considered vulnerable marine ecosystems (VMEs). Since many deep‐sea regions remain largely unexplored, investigating the occurrence of VME taxa on seamounts is challenging. Our study aimed to predict the distribution of four cold‐water coral (CWC) taxa, indicators for VMEs, in a region where occurrence data are scarce. Location: Seamounts around the Cabo Verde archipelago (NW Africa). Methods: We used species presence–absence data obtained from remotely operated vehicle (ROV) footage collected during two research expeditions. Terrain variables calculated using a multiscale approach from a 100‐m‐resolution bathymetry grid, as well as physical oceanographical data from the VIKING20X model, at a native resolution of 1/20°, were used as environmental predictors. Two modelling techniques (generalized additive model and random forest) were employed and single‐model predictions were combined into a final weighted‐average ensemble model. Model performance was validated using different metrics through cross‐validation. Results: Terrain orientation, at broad scale, presented one of the highest relative variable contributions to the distribution models of all CWC taxa, suggesting that hydrodynamic–topographic interactions on the seamounts could benefit CWCs by maximizing food supply. However, changes at finer scales in terrain morphology and bottom salinity were important for driving differences in the distribution of specific CWCs. The ensemble model predicted the presence of VME taxa on all seamounts and consistently achieved the highest performance metrics, outperforming individual models. Nonetheless, model extrapolation and uncertainty, measured as the coefficient of variation, were high, particularly, in least surveyed areas across seamounts, highlighting the need to collect more data in future surveys. Main Conclusions: Our study shows how data‐poor areas may be assessed for the likelihood of VMEs and provides important information to guide future research in Cabo Verde, which is fundamental to advise ongoing conservation planning. [ABSTRACT FROM AUTHOR]

Aim: Tidal marsh ecosystems are heavily impacted by human activities, highlighting a pressing need to address gaps in our knowledge of their distribution. To better understand the global distribution and changes in tidal marsh extent, and identify opportunities for their conservation and restoration, it is critical to develop a spatial knowledge base of their global occurrence. Here, we develop a globally consistent tidal marsh distribution map for the year 2020 at 10‐m resolution. Location: Global. Time period: 2020. Major taxa studied: Tidal marshes. Methods: To map the location of the world's tidal marshes at 10‐m resolution, we applied a random forest classification model to Earth observation data from the year 2020. We trained the classification model with a reference dataset developed to support distribution mapping of coastal ecosystems, and predicted the spatial distribution of tidal marshes between 60° N and 60° S. We validated the tidal marsh map using standard accuracy assessment methods, with our final map having an overall accuracy score of 0.85. Results: We estimate the global extent of tidal marshes in 2020 to be 52,880 km2 (95% CI: 32,030 to 59,780 km2) distributed across 120 countries and territories. Tidal marsh distribution is centred in temperate and Arctic regions, with nearly half of the global extent of tidal marshes occurring in the temperate Northern Atlantic (45%) region. At the national scale, over a third of the global extent (18,510 km2; CI: 11,200–20,900) occurs within the USA. Main conclusions: Our analysis provides the most detailed spatial data on global tidal marsh distribution to date and shows that tidal marshes occur in more countries and across a greater proportion of the world's coastline than previous mapping studies. Our map fills a major knowledge gap regarding the distribution of the world's coastal ecosystems and provides the baseline needed for measuring changes in tidal marsh extent and estimating their value in terms of ecosystem services. [ABSTRACT FROM AUTHOR]

Different turbulent entrainment‐mixing mechanisms between clouds and environment are essential to cloud‐related processes; however, accurate representation of entrainment‐mixing in weather/climate models still poses a challenge. This study exploits the use of machine learning (ML) to address this challenge. Four ML (Light Gradient Boosting Machine [LGB], eXtreme Gradient Boosting, Random Forest, and Support Vector Regression) are examined and compared. It is found that LGB performs best, and thus is selected to understand the impact of entrainment‐mixing on microphysics using simulation data from Explicit Mixing Parcel Model. Compared with traditional parameterizations, the trained LGB provides more accurate microphysical properties (number concentration and cloud droplet spectral dispersion). The partial dependences of predicted microphysics on features exhibit a strong alignment with physical mechanisms and expectations, as determined by the interpreting method, thus overcoming the limitations of the "black box" scheme. The underlying mechanisms are that the smaller number concentration and larger spectral dispersion correspond to more inhomogeneous entrainment‐mixing. Specifically, number concentration after entrainment‐mixing is positively correlated with adiabatic number concentration and liquid water content affected by entrainment‐mixing, and inversely correlated with adiabatic volume mean radius. Spectral dispersion after entrainment‐mixing is negatively correlated with liquid water content affected by entrainment‐mixing, turbulent dissipation rate and relative humidity of entrained air. Sensitivity analysis further suggests that number concentration is mainly determined by cloud microphysical properties whereas spectral dispersion is influenced by both cloud microphysical properties and environmental variables. The results indicate that the LGB scheme has the potential to enhance the representation of entrainment‐mixing in weather/climate models. Plain Language Summary: Entrainment‐mixing processes occurring between clouds and environmental air have significant effects on cloud‐climate feedback, precipitation, and radiative transfer. Accurately representing these processes has been challenging with previously proposed parameterizations. Machine learning (ML) excels at identifying complex nonlinear relationships and avoids the inherent limitations of conventional parameterizations. Thus, we explore the use of ML to acquire the number concentration and cloud droplet spectral dispersion affected by entrainment‐mixing processes. Simulation data from Explicit Mixing Parcel Model are employed to train, validate, and test the four MLs, including Light Gradient Boosting Machine (LGB), eXtreme Gradient Boosting, Random Forest and Support Vector Regression. After evaluation, the LGB is shown to obtain the most accurate microphysics among the four ML schemes and traditional parameterizations. Additionally, the interpreting method for peeking inside the "black box" reveals that a smaller number concentration and a larger spectral dispersion are indicative of the more inhomogeneous entrainment‐mixing. The relative correlations between predicted microphysics and various features align with our expectations and the underlying physical principles. Sensitivity tests further confirm that incorporating more features produces a more robust and efficient prediction. Overall, this study affirms the reliability and applicability of ML to develop/replace subgrid parameterizations in actual weather/climate models. Key Points: Machine learning schemes are trained to predict cloud microphysical properties affected by turbulent entrainment‐mixing processesThe proposed Light Gradient Boosting Machine provides more accurate microphysical properties compared with traditional parameterization schemesThe partial dependencies of microphysics on features prove a robust alignment with physical mechanisms through the interpreting method [ABSTRACT FROM AUTHOR]

Objective(s): An olfactory perceptual fingerprint (OPF) defines one's olfactory perception using perceptual descriptor ratings (such as odor pleasantness, intensity) for a set of odors. OPFs have been shown to distinguish patients with COVID‐related olfactory dysfunction (OD) and healthy controls with 86% accuracy. However, all participants rated the same odorants. With the aim to evaluate whether the OPFs are indeed odorant independent, previously published dataset by Lötsch et al. was reanalyzed. Furthermore, this independent dataset was used to check whether the OPFs separate patients with OD due to various causes from controls. Methods: The study included 104 controls and 42 patients, who were randomized into four odor sets with 10 odorants each. Odorants were presented using a computer‐controlled olfactometer and evaluated on scales from 1 (not at all) to 5 (very) using perceptual descriptors pleasant, intensive, familiar, edible, irritating, cold/warm, and painful. Results: Permutational multivariate analysis of variance showed that the odor set did not have a significant effect on the OPFs, confirming that the OPFs are indeed odorant independent. On the other hand, both diagnosis and age affected the OPFs (p <.001) and explained around 11% and 5% of the variance of the OPFs, respectively. Furthermore, a supervised machine learning method, random forest classifier, showed that OPF can distinguish patients and controls with 80% accuracy. Conclusion: OPFs are odorant independent. Patients perceived odors as less familiar, less intense, and less edible than controls. Other perceptual descriptors were much less important for the separation of patients and controls. Level of evidence: 3 [ABSTRACT FROM AUTHOR]

Background: Gastric cancer (GC), particularly for advanced stage of GC, commonly undergoes peritoneal metastasis (PM), which is the leading cause of GC‐related death. However, there currently has no reliable biomarker to predict the onset of GCPM. It is well known that the imbalance of gut microbiota contributes to the development and metastasis of gastrointestinal tumors. Unfortunately, little is known about how the alternation in gut microbiota is associated with the onset of GCPM. Methods: Our current study analyzed structural characteristics and functional prediction of gut microbiota in GC patients with PM (PM group) and without PM (non‐PM group). Fresh fecal samples were collected from a discovery cohort (PM = 38, non‐PM = 54) and a validation cohort (PM = 15, non‐PM = 21) of GC patients and their 16S ribosomal RNA (16s rRNA) gene amplicons were sequenced, followed by bioinformatics. Results: The results indicated an increase in the biodiversity of gut microbiota in the non‐PM group of the discovery cohort, compared with the PM group. Moreover, LEfSe analysis found 31 significantly different microorganisms, of which the Roseburia ranked the fifth in the random forest (RF) model. The characteristics of intestinal microbiota in GCPM patients were changed, and the abundance of Roseburia in gut microbiota from the GCPM patients was reduced and receiver operating characteristic (ROC) analysis revealed that the reduced abundance of gut Roseburia effectively predicted the onset of GCPM. Conclusion: This signature was also observed in the validation cohort. Therefore, Roseburia is a protective microbial marker and the reduced abundance of Roseburia in gut microbiota may help early diagnosis of GCPM. [ABSTRACT FROM AUTHOR]

Despite hosting ∼16% of the global freshwater and almost 50% of water resources in South America, Brazilian catchment‐scale relationships between drivers and streamflow are still poorly understood. Here, we used streamflow signatures and attributes of 735 catchments from the Catchment Attributes for Brazil data set to investigate the dominant hydrological processes for the catchments. We also assess how catchments group based on hydrologic behavior similarities and analyze which climatic/landscape attributes control the streamflow variability. To classify and group the catchments, we used the k‐means method optimized by the Elbow approach, along with a Principal Component Analysis. Uncertainty on catchment grouping was checked by k‐fold cross‐validation. Then, we used a recursive feature elimination using the random forest technique to assess the most influential catchment attributes to the hydrological signatures. Our results revealed six similarity groups, which followed mainly an aridity gradient ranging from the wettest to the driest, but also seasonality. The climate is the primary driver of hydrological behavior for the water‐limited groups, highlighting the influence and importance of the atmospheric demand in several Brazilian catchments. High soil storage capacity in energy‐limited catchments associated with high precipitation led to high discharge all year due to the subsurface fluxes' contribution. Our findings may be useful to improve streamflow predictability and hydrological behavior identification by further understanding hydrological similarities and their signatures due to catchment landscape characteristics. Further, by employing an easily reproducible methodology and clear metrics to weigh uncertainty, our study provides a significant step toward establishing a catchment‐scale common classification system. Key Points: We analyzed catchment‐scale relationships between streamflow and its drivers by clustering the catchments based on hydrological similaritySix catchment groups emerged following mainly an aridity gradient, from the wettest to the driest, and seasonalityClimate controls the hydrological behavior of water‐limited catchments, while landscape characteristics control energy‐limited catchments [ABSTRACT FROM AUTHOR]

Knowing where and when rivers cease to flow provides an important basis for evaluating riverine biodiversity, biogeochemistry and ecosystem services. We present a novel modeling approach to estimate monthly time series of streamflow intermittence at high spatial resolution at the continental scale. Streamflow intermittence is quantified at more than 1.5 million river reaches in Europe as the number of no‐flow days grouped into five classes (0, 1–5, 6–15, 16–29, 30–31 no‐flow days) for each month from 1981 to 2019. Daily time series of observed streamflow at 3706 gauging stations were used to train and validate a two‐step random forest modeling approach. Important predictors were derived from time series of monthly streamflow at 73 million 15 arc‐sec (∼500 m) grid cells that were computed by downscaling the 0.5 arc‐deg (∼55 km) output of the global hydrological model WaterGAP, which accounts for human water use. Of the observed perennial and non‐perennial station‐months, 97.8% and 86.4%, respectively, were correctly predicted. Interannual variations of the number of non‐perennial months at non‐perennial reaches were satisfactorily simulated, with a median Pearson correlation of 0.5. While the spatial prevalence of non‐perennial reaches is underestimated, the number of non‐perennial months is overestimated in dry regions of Europe where artificial storage abounds. Our model estimates that 3.8% of all European reach‐months and 17.2% of all reaches were non‐perennial during 1981–2019, predominantly with 30–31 no‐flow days. Although estimation uncertainty is high, our study provides, for the first time, information on the continent‐wide dynamics of non‐perennial rivers and streams. Plain Language Summary: Even in wet climates, small streams can seasonally dry up. In drier areas, large rivers might not carry water for weeks or months. However, as streamflow observations are lacking for most drying rivers, we know little about when, where, and how long rivers experience such a streamflow intermittence that is crucial for both river life and human water supply. We developed and applied a novel approach to estimate, for the first time, the temporal dynamics of streamflow intermittence across European rivers and streams, including small ones. This approach combines the output of a global hydrological model with streamflow observations and other data. We refined the global model output available for 50 km cells to monthly streamflow in 500 m cells. We then applied a machine learning model to predict the number of days without water flow in each month during the period 1981–2019 for over 1.5 million river segments. We found that 17% of all European segments and 4% of all months at all segments experienced at least one day without flow. In the future, the model will be used to estimate the impact of climate change on streamflow intermittence. Key Points: Streamflow intermittence at more than 1.5 million European reaches was estimated for every month during 1981–201918.7% of the European river network length and 3.8% of all reach‐months are non‐perennial, predominantly with 30–31 no‐flow days15 arc‐sec monthly streamflow obtained by downscaling the output of a global hydrological model serves as input to random forest modeling [ABSTRACT FROM AUTHOR]

Horticultural peat extraction can mobilize dissolved organic matter (DOM) and inorganic nutrients (nitrogen and phosphorous) to surface waters, harming aquatic ecosystems and water quality. However, it is uncertain how peat extraction affects solute concentration across hydrological and seasonal conditions and how biogeochemical processing in downstream drainage networks responds. Over two years, we used repeated, spatially extensive sampling in stream networks of two mixed land‐use catchments (<200 km2) on the subhumid interior plains of western Canada. We used random forest models to disentangle the effects of land cover, hydrology, and temperature on water chemistry. Peatlands were the dominant source of DOM to streams, but we detected no substantial effect of peat extraction on DOM concentration or composition. Stream discharge was the most important predictor of DOM composition, with generally humic‐like DOM becoming fresher during snowmelt and summer base flow. We detected no effect from peat extraction on soluble reactive phosphorous (SRP) or nitrate (NO3−). However, total ammonia nitrogen (TAN) was an order of magnitude higher in subcatchments with >40% extracted peatland cover (median: 1.5 mg TAN L−1) compared to catchments with similar intact peatland cover. Mass balance analysis suggested that DOM and inorganic nutrients synchronously attenuated during low flows. During high flows, DOM and inorganic nitrogen were conservatively transported, while SRP was attenuated, likely sorbing to suspended particles. Our study suggests that excess TAN mobilized by peat extraction is utilized in headwaters during low flow but propagates downstream during high flow, with implications for eutrophication that land managers should consider. Key Points: Peat extraction in a subhumid continental climate significantly increased stream ammonia concentration, but not carbon or phosphorusStream hydrology strongly influenced dissolved organic matter (DOM) composition, with low flows shifting DOM from aromatic to aliphaticHydrology affected the downstream persistence of DOM and nitrogen, which attenuated at low flows. Phosphorus was attenuated at all flows [ABSTRACT FROM AUTHOR]

Numerous factors influence mountain biodiversity variation across elevational gradients and recognizing the relative importance is vital for understanding species distribution mechanisms. We examined oribatid mites at nine elevations (from 600 to 2200 m a.s.l) and four vegetation types from mixed coniferous and broad‐leaved forests to alpine tundra on Changbai Mountain. We assessed the contribution of environmental factors (climatic and local factors) and spatial processes (geographic or elevation distances) to oribatid mite community assembly and identified 59 oribatid mite species from 38 families and 51 genera. With increasing elevation, species richness and the Shannon index declined significantly, whereas abundance followed a hump‐shaped trend. Soil TP, NH4+‐N, MAT, MAP, and elevation were the critical variables shaping oribatid mite communities based on random forest analysis. Moreover, environmental and spatial factors, and oribatid mite communities were significantly correlated based on Mantel and partial Mantel tests. Local characteristics (3.9%), climatic factors (1.9%), and spatial filtering (8.8%) played crucial roles in determining oribatid mite communities across nine elevational bands (based on variation partitioning analyses of abundance data). Within the same vegetation types, spatial processes had relatively little effects, with local characteristics the dominant drivers of oribatid mite community variation. Environmental and spatial filters together shape oribatid mite community assembly and their relative roles varied with elevation and vegetation type. These findings are crucial for the conservation, restoration, and management of Changbai mountain ecosystems in the context of climate change, along with the prediction of future vertical biotic gradient pattern evolution. [ABSTRACT FROM AUTHOR]

Assessing the extinction risk of species based on the International Union for Conservation of Nature (IUCN) Red List (RL) is key to guiding conservation policies and reducing biodiversity loss. This process is resource demanding, however, and requires continuous updating, which becomes increasingly difficult as new species are added to the RL. Automatic methods, such as comparative analyses used to predict species RL category, can be an efficient alternative to keep assessments up to date. Using amphibians as a study group, we predicted which species are more likely to change their RL category and thus should be prioritized for reassessment. We used species biological traits, environmental variables, and proxies of climate and land‐use change as predictors of RL category. We produced an ensemble prediction of IUCN RL category for each species by combining 4 different model algorithms: cumulative link models, phylogenetic generalized least squares, random forests, and neural networks. By comparing RL categories with the ensemble prediction and accounting for uncertainty among model algorithms, we identified species that should be prioritized for future reassessment based on the mismatch between predicted and observed values. The most important predicting variables across models were species’ range size and spatial configuration of the range, biological traits, climate change, and land‐use change. We compared our proposed prioritization index and the predicted RL changes with independent IUCN RL reassessments and found high performance of both the prioritization and the predicted directionality of changes in RL categories. Ensemble modeling of RL category is a promising tool for prioritizing species for reassessment while accounting for models’ uncertainty. This approach is broadly applicable to all taxa on the IUCN RL and to regional and national assessments and may improve allocation of the limited human and economic resources available to maintain an up‐to‐date IUCN RL. [ABSTRACT FROM AUTHOR]

Objective: The aim of this study was to develop a machine learning algorithm using an off‐the‐shelf digital watch, the Samsung watch (SM‐R800), and evaluate its effectiveness for the detection of generalized convulsive seizures (GCS) in persons with epilepsy. Methods: This multisite epilepsy monitoring unit (EMU) phase 2 study included 36 adult patients. Each patient wore a Samsung watch that contained accelerometer, gyroscope, and photoplethysmographic sensors. Sixty‐eight time and frequency domain features were extracted from the sensor data and were used to train a random forest algorithm. A testing framework was developed that would better reflect the EMU setting, consisting of (1) leave‐one‐patient‐out cross‐validation (LOPO CV) on GCS patients, (2) false alarm rate (FAR) testing on nonseizure patients, and (3) "fixed‐and‐frozen" prospective testing on a prospective patient cohort. Balanced accuracy, precision, sensitivity, and FAR were used to quantify the performance of the algorithm. Seizure onsets and offsets were determined by using video‐electroencephalographic (EEG) monitoring. Feature importance was calculated as the mean decrease in Gini impurity during the LOPO CV testing. Results: LOPO CV results showed balanced accuracy of.93 (95% confidence interval [CI] =.8–.98), precision of.68 (95% CI =.46–.85), sensitivity of.87 (95% CI =.62–.96), and FAR of.21/24 h (interquartile range [IQR] = 0–.90). Testing the algorithm on patients without seizure resulted in an FAR of.28/24 h (IQR = 0–.61). During the "fixed‐and‐frozen" prospective testing, two patients had three GCS, which were detected by the algorithm, while generating an FAR of.25/24 h (IQR = 0–.89). Feature importance showed that heart rate‐based features outperformed accelerometer/gyroscope‐based features. Significance: Commercially available wearable digital watches that reliably detect GCS, with minimum false alarm rates, may overcome usage adoption and other limitations of custom‐built devices. Contingent on the outcomes of a prospective phase 3 study, such devices have the potential to provide non‐EEG‐based seizure surveillance and forecasting in the clinical setting. [ABSTRACT FROM AUTHOR]

Convective initiation (CI) nowcasting in subtropical regions often faces challenges, such as complex physical processes and imbalanced samples of CI events, resulting in a high false alarm ratio (FAR). In this paper, we propose a Storm Warning System with Physics‐Augmentation (SWASP) based on the random forest algorithm and cloud physical conditions, using Himawari‐8 Advanced Himawari Imager data from April to September 2019 in South China. The cloud physical conditions (e.g., cloud‐top cooling rates) were investigated to establish regional thresholds for convection occurrence. Ancillary information, including elevation, satellite zenith angle, and latitude, was also incorporated into the SWASP model. Compared to conventional methods, the SWASP model exhibits an improved probability of detection by 0.11 and 0.08 and a decreased FAR by 0.38 and 0.44 for daytime and nighttime forecasts. Moreover, the SWASP model enables the detection of local convective storm systems about 30 min to 1 hr ahead of radar detection in typical convective storm cases. This study contributes to further advancements of the SWASP model by incorporating physical conditions and emphasizes the potential application of geostationary satellites in convective early warnings. Key Points: A Storm Warning System with Physics‐Augmentation (SWASP) has been proposed, which utilizes the random forest algorithmSWASP enhances the probability of detection and reduces false alarm ratios in storm nowcastingChanges in cloud‐top height are identified as the critical factor for accurate storm nowcasting [ABSTRACT FROM AUTHOR]

Understanding preferential flow (PF) at large scales is critical for improving land management and groundwater (GW) quality. However, limited knowledge of this process, due to soil surface heterogeneity and observational constraints, hampers progress. In this study, we propose estimating effective PF at remote sensing footprint scale (4–9 km) by examining its impact on soil moisture (SM) distribution and shallow groundwater (SGW) table fluctuations (depth ≤$ \le $ 5 m). Effective PF encompasses macropore, funnel, and finger flow pathways influencing SGW table fluctuations. We compiled daily SGW observations (2019–2021) from 19 Continental United States (CONUS) sites through United States Geological Survey. Using inverse modeling in HYDRUS‐1D, SGW data, and climate hazards group infrared precipitation with station data precipitation, we inversely estimated soil hydraulic parameters of the dual‐porosity model (DPM) simulating vertical flow from soil surface to subsurface. Effective PF presence was inferred using three criteria: (1) daily precipitation equal to or exceeding the site‐specific average across multiple (calibration) years, (2) daily observed SGW table increase, and (3) daily difference between observed and DPM simulated SGW tables ≤$ \le $ 50% of the site‐specific root mean square error. Leveraging optimized DPM parameters and associated soil texture, classified PF events, and soil moisture active passive (SMAP L3E) satellite‐based SM, a random forest algorithm with 10‐fold cross validation predicted large‐scale effective PF events. Results indicate seasonal dependence, with spring having the highest occurrence of PF events. The random forest model achieved 98% accuracy in predicting large‐scale PF events, with SMAP SM and saturated hydraulic conductivity (Ks) among the four most impactful variables. Our approach provides a soil hydraulic property, site characteristic, soil texture, and remote sensing‐based generalized tool to analyze large‐scale effective PF. Core Ideas: Soil hydraulic properties driving preferential flow were estimated across CONUS.Large‐scale effective preferential flow seasonally impacts shallow groundwater tables.Remotely sensed precipitation and soil moisture predicted large‐scale preferential flow well. [ABSTRACT FROM AUTHOR]

Machine learning (ML) has been broadly applied for vadose zone applications in recent years. This article provides a comprehensive review of such developments. ML applications for variables corresponding to different complex vadose zone processes are summarized mostly in a prediction context. By analyzing and assessing these applications, we discovered extensive usages of classic ML models with relatively limited applications of deep learning (DL) approaches in general. We also recognized a lack of benchmark datasets for soil property research as well as limited integration of physics‐based vadose zone principles into the ML approaches. To facilitate this interdisciplinary research of ML in vadose zone characterization and processes, a paradigm of knowledge‐guided machine learning is suggested along with other data‐driven and ML model‐based research suggestions to advance future research. Core Ideas: Random forest and artificial neural network are two widely applied machine learning options for predicting vadose zone studies.A benchmark dataset is missing in soil property studies.We suggest vadose zone scientists explore more deep learning options and expand knowledge‐guided machine learning implementations. [ABSTRACT FROM AUTHOR]

Snow density is of paramount importance in water resource management, snow avalanche warning, and climate change research. However, the lack of competent methods for long‐term and vast‐scale snow density mapping persists due to the intricate spatiotemporal dependencies inherent in snow density, resulting in scarce and inaccurate snow density products. To address this challenge, a spatiotemporal random forest (STRF) model is constructed by leveraging in‐situ measurements, multisource remote sensing, and reanalysis data. It tackles the spatiotemporal dependencies in snow density arising from its inherent heterogeneity and the relations involving snow density and nonlinearly connected meteorological, terrain, vegetation, and snow‐related factors. The effectiveness of the model is substantiated through rigorous validation methods, including random, temporal, and spatial block cross‐validations as well as independent validation, apparently surpassing ERA5‐Land snow density. The estimated snow density is also demonstrated to be able to improve existing snow water equivalent data set using fixed snow density. Utilizing the proposed model, a data set of daily 25‐km snow density from 1980 to 2018 is constructed for stable snow cover areas in China, which holds significant potential for research and applications in the realm of snow hydrology. Plain Language Summary: Snow density plays a crucial role in managing water resources and issuing snow avalanche warnings. Snow density exhibits great spatiotemporal dependent structure due to its spatiotemporal heterogeneity and its relations with various influencing factors, which pose challenges to mapping long‐term and large‐scale snow density. This leads to a scarcity of accurate snow density products. To tackle this issue, we develope a spatiotemporal random forest (STRF) model by combining ground measurements, remote sensing sources, and reanalysis data. Notably, our model shows impressive results apparently outperforming the ERA5‐Land snow density data set. Our estimated snow density can also enhance existing snow water equivalent data set that rely on fixed snow density. Using our model, we produce a data set of daily 25‐km snow density from 1980 to 2018 for stable snow cover areas in China. This data set holds significant potential for research and practical applications in the field of snow hydrology. Key Points: A spatiotemporal random forest (STRF) model is proposed for estimating large‐scale and long‐term snow densitySTRF depicts the spatiotemporal dependent structure of snow density and handles its nonlinear relations with various influencing factorsThe estimated snow density outperforms ERA5‐Land snow density and could improve snow water equivalent data set using fixed snow density [ABSTRACT FROM AUTHOR]

The lack of discharge observations and reliable drainage information is a pervasive problem in urban catchments, resulting in difficulties in parameterizing urban hydrological models. Current parameterization methods for ungauged urban catchments mostly rely on subjective experiences or simplified models, resulting in inadequate accuracy for urban flood prediction. Parameter regionalization has been widely used to tackle model parameterization issues, but has rarely been employed for urban hydrological models. How to conduct effective parameter regionalization for urban hydrological models remains to be investigated. Here we propose a parameter regionalization framework (PRF) that integrates donor catchment clustering and the optimal regression‐based methods in each cluster. The PRF is applied to an urban hydrological model, the Time Variant Gain Model in urban areas (TVGM_Urban), in 37 urban catchments in Shenzhen City, China. We first show satisfactory flood simulation performance of TVGM_Urban for all urban catchments. Subsequently, we employ the PRF for parameter regionalization of TVGM_Urban. PRF classifies 37 urban catchments into three groups, and the partial least‐squares regression is identified as optimal regression‐based method for Groups 1 and 2, while the random forest model is found to be best for Group 3. To evaluate the simulation performance of PRF, we compare it with eight single regionalization methods. The results indicate better simulation performance and lower uncertainty of PRF, and donor catchment clustering can effectively enhance the simulation performance of linear regression‐based methods. Lastly, we identify curve number, land cover area ratios, and slope as critical factors for most TVGM_Urban parameters based on PRF results. Plain Language Summary: Frequent urban flooding and waterlogging pose significant threats to human life and property. Accurate modeling urban floods is crucial for minimizing damages caused by urban flooding. However, the delayed initiation of urban observations has resulted in prevalent data scarcity in urban areas, leading to difficulties and low accuracy in estimating model parameters for ungauged urban catchments. This study proposes a parameter regionalization framework for an urban hydrological model to improve the accuracy of flood simulations in ungauged urban catchments. This framework involves classifying gauged catchments into multiple groups and identifying the optimal regression‐based methods in each group for parameter regionalization. The flood simulation performance and stability of the parameter regionalization framework is evaluated by comparing it with eight widely‐used regionalization methods in 37 urban catchments located in Shenzhen City, China. Our results show that the classification of gauged catchments can significantly improve the simulation performance of linear regression methods. Additionally, the parameter regionalization framework outperforms the other regionalization methods for both urban flood simulation in the study area and urban flood prediction in the ungauged urban catchments. This study provides a novel and efficient parameterization method for accurate flood prediction in ungauged urban catchments to reduce potential losses. Key Points: TVGM_Urban presents satisfying performance in urban flood modelingThe PRF integrating donor catchment clustering and optimal regression methods in each cluster outperforms single regionalization methodsCurve number, land cover area ratio, and slope are critical factors for runoff generation parameters of TVGM_Urban [ABSTRACT FROM AUTHOR]

Inter‐catchment groundwater flow (IGF) plays an essential role in streamflow generation and water quality in forested headwaters. Multiple factors are thought to contribute to IGF, including climate, topographical, and geological factors. However, studies have not clarified the relationships between IGF and catchment properties in the headwater catchments due to the lack of observational data at scales smaller than 100 ha. This study examined possible factors influencing IGF using random forest analysis based on annual water balance data from 152 forested catchments ranging from 0.09 to 9400 ha in Japan. The results showed that catchment scale had the greatest influence on IGF, and IGF tended to decrease with increasing catchment area at scales of less than 10 ha. The average IGF stabilized around zero in catchments greater than 10 ha. The averaged IGF trend with catchment scale indicated more outward groundwater flow in catchments smaller than 10 ha, but no relationship between IGF and catchment size in catchments larger than 10 ha. The variability in IGF decreased with catchment size and was lowest at 10–100 ha. The decrease in variability in catchments less than 100 ha was mainly due to river confluence and the increased variability in catchments larger than 100 ha indicated potential observation errors increase in catchments of this size. Plain Language Summary: Inter‐catchment groundwater flow (IGF) refers to groundwater flux across surface topographic boundaries. Recent studies have clarified that IGF significantly affects streamflow generation and water quality. However, direct observation of IGF is difficult, and the mechanisms underlying IGF are not fully understood. This study examined the factors influencing IGF, as well as its scale dependence and variability based on water balance data from 152 forested catchments ranging from 0.09 to 9400 ha in Japan. This showed that catchment area had the greatest influence on IGF. Groundwater tended to flow out of catchments smaller than 10 ha, and stabilized in those larger than 10 ha. Variability in the IGF decreased with catchment size at less than 100 ha. These results suggest that on the scale of hillslope and headwater catchments, IGF has a great influence on streamflow generation. Key Points: We examined the factors influencing inter‐catchment groundwater flow (IGF) using water balance data for 152 forest catchments in JapanCatchment scale was the factor with the greatest influence on the IGFGroundwater tends to flow out of catchments smaller than 10 ha and IGF approaches zero when the catchment area exceeds 10 ha [ABSTRACT FROM AUTHOR]

Significant and readily accessible orogenic gold deposits have been previously recognized, exploited, and progressively depleted. Innovative approaches are required to discover new and deeply buried deposits. Recently, trace element variations in apatite have been used to distinguish fertile and barren environments as reliable mineral exploration tools. In this study, machine learning models using Random Forest and Deep Neutral Network are utilized to assess the fertility of quartz veins and altered zones in the orogenic gold systems. The two models have been trained using trace element data of apatite, and the performance of both models yield good classification accuracy (∼90% on average) with low false positive rates. Feature importance analysis (Gini decrease and hidden layer weights) suggest that Pb, As, U, Sr, Eu, Mn, and Fe are the important parameters. Arsenic, U, Eu, Mn, and Fe are redox‐sensitive elements, with their concentrations responding to changes in fluid redox conditions. Strontium primarily originates from the breakdown of plagioclase, which is more likely to occur under oxidizing fluid conditions. Therefore, we infer that the main controlling factor of the models is the redox conditions. The two distinct models consistently highlight the most significant contribution of Pb to this differentiation, even though Pb is not a redox‐sensitive element and can only substitute for Ca2+ in apatite as Pb2+. We infer that the high contribution of Pb may be attributed to the potential transportation of Au in the form of a Pb‐(Bi)‐Au melt, and the Pb content in apatite is influenced by the Pb content in the melt, fluid oxygen, and sulfur fugacity. We also propose a novel discriminant plot using Linear Discriminant Analysis (LDA) to assess the mineralization potential in quartz veins and alteration zones based on apatite trace element data. The machine learning and LDA results suggest that apatite trace elements could be used effectively in the future orogenic gold deposit exploration. Key Points: Random Forests and Deep Neutral Network models demonstrate effective discrimination among various apatite types in orogenic gold depositsThe two distinct models consistently highlight the significant contribution of Pb to this differentiationA novel discriminant plot using Linear Discriminant Analysis (LDA) is proposed to assess the mineralization potential [ABSTRACT FROM AUTHOR]

Understanding the processes and mechanisms that shape the distribution patterns and variations of biodiversity along spatial gradients continues to be a priority for ecological research. We focused on the biodiversity of benthic diatom communities within a large near‐natural watershed. The objectives are: (1) to explore the overall spatial patterns of benthic diatom biodiversity; (2) to investigate the effects associated with watercourse position and environmental variables, as well as both common and rare species on two facets (i.e., taxonomic and functional) of alpha and beta diversity; and (3) to unveil the mechanisms underlying their spatial variations. Alpha diversity indices along the stream watercourse showed a clear increasing trend from upstream to downstream sites. Results of random forest regression identified conductivity as the primary factor influencing functional alpha diversity, while elevation emerged as the predominant factor for taxonomic alpha diversity. Beta diversity partitioning revealed that taxonomic beta diversity generally exceeded functional beta diversity. These diversity measures exhibited different patterns along the watercourse position: taxonomic beta diversity remained relatively consistent along the watercourse, whereas functional total beta diversity and its two components of middle stream sites were lower than those of upstream and downstream sites. Functional beta diversity was sustained by dominant and common species, while rare species made significant contributions to taxonomic beta diversity. Both taxonomic and functional beta diversity and its components displayed a stronger influence from spatial factors than from local environmental, geo‐climatic, and nutrient variables. Collectively, taxonomic and functional alpha and beta diversity demonstrated distinct responses to the main environmental gradients and spatial factors within our catchment, highlighting their different insights into diatom diversity. Furthermore, research is required to assess the generalizability of our findings to similar ecosystems. In addition, this study presents opportunities for expansion to include other taxa (e.g., macroinvertebrates and fish) to gain a comprehensive understanding of the driving mechanisms behind stream biodiversity. [ABSTRACT FROM AUTHOR]

To conserve wide‐ranging species in degraded landscapes, it is essential to understand how the behavior of animals changes in relation to the degree and composition of modification. Evidence suggests that large inter‐individual variation exists in the propensity for use of degraded areas and may be driven by both behavioral and landscape factors. The use of cultivated lands by wildlife is of particular interest, given the importance of reducing human‐wildlife conflicts and understanding how such areas can function as biodiversity buffers. African elephant space use can be highly influenced by human activity and the degree to which individuals crop‐raid. We analyzed GPS data from 56 free‐ranging elephants in the Serengeti‐Mara Ecosystem using resource selection functions (RSFs) to assess how crop use may drive patterns of resource selection and space use within a population. We quantified drivers of similarity in resource selection across individuals using proximity analysis of individual RSF coefficients derived from random forest models. We found wide variation in RSF coefficient values between individuals indicating strongly differentiated resource selection strategies. Proximity assessment indicated the degree of crop use in the dry season, individual repeatability, and time spent in unprotected areas drove similarity in resource selection patterns. Crop selection was also spatially structured in relation to agricultural fragmentation. In areas with low fragmentation, elephants spent less time in crops and selected most strongly for crops further from protected area boundaries, but in areas of high fragmentation, elephants spent twice as much time in crops and selected most strongly for crops closer to the protected area boundary. Our results highlight how individual differences and landscape structure can shape use of agricultural landscapes. We discuss our findings in respect to the conservation challenges of human‐elephant conflict and incorporating behavioral variation into human‐wildlife coexistence efforts. [ABSTRACT FROM AUTHOR]

We combined two climate‐based distribution models with three finer‐scale suitability models to identify habitat for pronghorn recovery in California now and into the future. We used a consensus approach to identify areas of suitable climate now and future for pronghorn in California. We compared the results of climate models from two separate hypotheses about their historical ecology in the state. Under the migration hypothesis, pronghorn were expected to be limited climatically by extreme cold in winter and extreme heat in summer; under the niche reduction hypothesis, historical pronghorn of distribution would have better represented the climatic limitations of the species. We combined occurrences from GPS collars distributed across three populations of pronghorn in the state to create three distinct habitat suitability models: (1) an ensemble model using random forests, Maxent, classification and regression Trees, and a generalized linear model; (2) a step selection function; and (3) an expert‐driven model. We evaluated consensus among both the climate models and the suitability models to prioritize areas for, and evaluate the prospects of, pronghorn recovery. Climate suitability for pronghorn in the future depends heavily on model assumptions. Under the migration hypothesis, our model predicted that there will be no suitable climate in California in the future. Under the niche reduction hypothesis, by contrast, suitable climate will expand. Habitat suitability also depended on the methods used, but areas of consensus among all three models exist in large patches throughout the state. Identifying habitat for a species which has undergone extreme range collapse, and which has very fine scale habitat needs, presents novel challenges for spatial ecologists. Our multimethod, multihypothesis approach can allow habitat modelers to identify areas of consensus and, perhaps more importantly, fill critical knowledge gaps that could resolve disagreements among the models. For pronghorn, a better understanding of their upper thermal tolerances and whether historical populations migrated will be crucial to their potential recovery in California and throughout the arid Southwest. [ABSTRACT FROM AUTHOR]

Summary: Aims: A multi‐stakeholder consensus has proposed MASLD (metabolic dysfunction‐associated steatotic liver disease). We aimed to investigate the pathological findings related to the mid‐term mortality of patients with biopsy‐proven MASLD in Japan. Methods: We enrolled 1349 patients with biopsy‐proven MASLD. The observational period was 8010 person years. We evaluated independent factors associated with mortality in patients with MASLD by Cox regression analysis. We also investigated pathological profiles related to mortality in patients with MASLD using data‐mining analysis. Results: The prevalence of MASH and stage 3/4 fibrosis was observed in 65.6% and 17.4%, respectively. Forty‐five patients with MASLD died. Of these, liver‐related events were the most common cause at 40% (n = 18), followed by extrahepatic malignancies at 26.7% (n = 12). Grade 2/3 lobular inflammation and stage 3/4 fibrosis had a 1.9‐fold and 1.8‐fold risk of mortality, respectively. In the decision‐tree analysis, the profiles with the worst prognosis were characterised by Grade 2/3 hepatic inflammation, along with advanced ballooning (grade 1/2) and fibrosis (stage 3/4). This profile showed a mortality at 8.3%. Furthermore, the random forest analysis identified that hepatic fibrosis and inflammation were the first and second responsible factors for the mid‐term prognosis of patients with MASLD. Conclusions: In patients with biopsy‐proven MASLD, the prevalence of MASH and advanced fibrosis was approximately 65% and 20%, respectively. The leading cause of mortality was liver‐related events. Hepatic inflammation and fibrosis were significant factors influencing mid‐term mortality. These findings highlight the importance of targeting inflammation and fibrosis in the management of patients with MASLD. [ABSTRACT FROM AUTHOR]

Selective breeding for production traits has yielded relatively rapid successes with high‐fecundity aquaculture species. Discovering the genetic changes associated with selection is an important goal for understanding adaptation and can also facilitate better predictions about the likely fitness of selected strains if they escape aquaculture farms. Here, we hypothesize domestication as a genetic change induced by inadvertent selection in culture. Our premise is that standardized culture protocols generate parallel domestication effects across independent strains. Using eastern oyster as a model and a newly developed 600K SNP array, this study tested for parallel domestication effects in multiple independent selection lines compared with their progenitor wild populations. A single contrast was made between pooled selected strains (1–17 generations in culture) and all wild progenitor samples combined. Population structure analysis indicated rank order levels of differentiation as [wild − wild] < [wild − cultured] < [cultured − cultured]. A genome scan for parallel adaptation to the captive environment applied two methodologically distinct outlier tests to the wild versus selected strain contrast and identified a total of 1174 candidate SNPs. Contrasting wild versus selected strains revealed the early evolutionary consequences of domestication in terms of genomic differentiation, standing genetic diversity, effective population size, relatedness, runs of homozygosity profiles, and genome‐wide linkage disequilibrium patterns. Random Forest was used to identify 37 outlier SNPs that had the greatest discriminatory power between bulked wild and selected oysters. The outlier SNPs were in genes enriched for cytoskeletal functions, hinting at possible traits under inadvertent selection during larval culture or pediveliger setting at high density. This study documents rapid genomic changes stemming from hatchery‐based cultivation of eastern oysters, identifies candidate loci responding to domestication in parallel among independent aquaculture strains, and provides potentially useful genomic resources for monitoring interbreeding between farm and wild oysters. [ABSTRACT FROM AUTHOR]

Cost‐sensitive ensemble learning as a combination of two approaches, ensemble learning and cost‐sensitive learning, enables generation of cost‐sensitive tree‐based ensemble models using the cost‐sensitive decision tree (CSDT) learning algorithm. In general, tree‐based models characterize nice graphical representation that can explain a model's decision‐making process. However, the depth of the tree and the number of base models in the ensemble can be a limiting factor in comprehending the model's decision for each sample. The CSDT models are widely used in finance (e.g., credit scoring and fraud detection) but lack effective explanation methods. We previously addressed this gap with cost‐sensitive tree Shapley Additive Explanation Method (CSTreeSHAP), a cost‐sensitive tree explanation method for the single‐tree CSDT model. Here, we extend the introduced methodology to cost‐sensitive ensemble models, particularly cost‐sensitive random forest models. The paper details the theoretical foundation and implementation details of CSTreeSHAP for both single CSDT and ensemble models. The usefulness of the proposed method is demonstrated by providing explanations for single and ensemble CSDT models trained on well‐known benchmark credit scoring datasets. Finally, we apply our methodology and analyze the stability of explanations for those models compared to the cost‐insensitive tree‐based models. Our analysis reveals statistically significant differences between SHAP values despite seemingly similar global feature importance plots of the models. This highlights the value of our methodology as a comprehensive tool for explaining CSDT models. [ABSTRACT FROM AUTHOR]

Aim: To investigate the prevalence and clinical manifestations of reading, writing, and mathematics disorders in children with cerebral palsy (CP). We explored how the clinical profile of these children differed from those with specific learning disorders (SLDs), taking into account several factors, particularly IQ scores, neuropsychological aspects, and the presence of a visual impairment. Method: A prospective cross‐sectional study was conducted in 42 children with CP (mean age 9 years 8 months; SD = 2 years 2 months) and 60 children with SLDs (mean age 10 years; SD = 1 year 7 months). Clinical characteristics, neuromotor and cognitive profiles, neuropsychological aspects (speech performance, academic skills, visual attention, phonological awareness, working memory), and signs of visual impairment (visual acuity, contrast sensitivity, visual field, oculomotor functions) were assessed. A machine learning approach consisting of a random forest algorithm, where the outcome was the diagnosis and the covariates were the clinical variables collected in the sample, was used for the analyses. Results: About 59% of the children with CP had reading, writing, or mathematics disorders. Children with CP with learning disorders had a low performance IQ, normal phonological awareness, and working memory difficulties, whereas children with SLDs had normal performance IQ, impaired phonological awareness, and mild working memory difficulties. There were no differences in verbal IQ between the two groups. Interpretation: Learning disorders are frequently associated with CP, with different clinical characteristics, compared with SLDs. Assessment of academic skills is mandatory in these children, even if the IQ is normal. At school age, specific interventions to promote academic skills in children with CP could be a major rehabilitative goal. What this paper adds: Reading, writing, and mathematics disorders in cerebral palsy have specific clinical characteristics.Their underlying mechanisms differ from those described in specific learning disorders.Working memory impairment can be considered a hallmark of learning disorders in children with cerebral palsy. [ABSTRACT FROM AUTHOR]

Inversion of magnetic basement interfaces in basins is essential for interpreting potential field data and studying geothermal resource distribution, as well as basin formation and evolution. This paper introduces a novel method for inverting magnetic basement interfaces using a random forest regression (RFR) algorithm that combines potential field processing and machine learning techniques. The method creates magnetic base interface models and corresponding magnetic anomaly data through the random midpoint displacement method and magnetic interface finite element forward simulation. These anomalies are then processed using techniques such as directional transformations, analytical continuation, spatial derivatives, and fractional transformations. Feature attributes are extracted, and Gini importance is utilized to measure the contributions of feature factors, identify effective features, and enhance algorithm efficiency. The validity and practicality of the method are demonstrated through the analysis of both idealized and noisy models. The proposed machine learning‐based approach is more intelligent, efficient, and accurately represents the relief of magnetic base interfaces. When applied to magnetic survey data in the Datong Basin, it produced a reliable basin base model that aligns with known structural information, paving the way for further research in magnetic interface inversion. Key Points: A machine learning method based on random forest regression algorithm is used to recover the relief of the magnetic basement interface by using high‐precision magnetic dataApplying potential field processing and transformation to feature extraction for machine learningThe proposed technique intelligently and efficiently restores the 3D relief of the magnetic basement interface, and it has been successfully implemented in the Datong Basin's Tianzhen‐Yanggao sag [ABSTRACT FROM AUTHOR]

Background: The time‐series data of the Normalized Difference Vegetation Index (NDVI) is a crucial indicator for global and regional vegetation monitoring. However, the current assessment of global and regional long‐term vegetation changes is subject to large uncertainties due to the lack of spatiotemporally continuous time‐series data sets. Methods: In this study, a long time‐series monthly NDVI data set with a spatial resolution of 250 m from 1982 to 2020 was developed by combining Moderate Resolution Imaging Spectroradiometer (MODIS) and AVHRR (Advanced Very High‐Resolution Radiometer) time‐series NDVI products using the Random Forest (RF) downscaling model. Results: Compared to the MODIS NDVI product, the fused product shows RMSE and mean absolute error ranging from 0 to 0.075 and from 0 to 0.05, respectively, with R2 values mostly above 0.7. Conclusions: The long time‐series NDVI products generated in this study are reliable in terms of accuracy and have great potential for long‐term dynamic monitoring of terrestrial ecosystems on the Qinghai–Tibet Plateau. [ABSTRACT FROM AUTHOR]

Accurately predicting species' range shifts in response to environmental change is paramount for understanding ecological processes and global change. In synthetic analyses, traits emerge as significant but weak predictors of species' range shifts across recent climate change. These studies assume linear responses to traits, while detailed empirical work often reveals trait responses that are unimodal and contain thresholds or other nonlinearities. We hypothesize that the use of linear modeling approaches fails to capture these nonlinearities and, therefore, may be under‐powering traits to predict range shifts. We evaluate the predictive performance of approaches that can capture nonlinear relationships (ridge‐regularized linear regression, support vector regression with linear and nonlinear kernels, and random forests). We apply our models using six multidecadal range shift datasets for plants, moths, marine fish, birds, and small mammals. We show that nonlinear approaches can perform better than least‐squares linear modeling in reproducing historical range shifts. Consistent with expectations, we identify dispersal and climatic niche traits as primary determinants of distribution shifts. Traits identified as important predictors and the direction of trait effects are generally consistent across models, but there are notable exceptions. Among important predictors, there are more consistent responses to climatic niches than dispersal ability. Modest improvements in predictability when accounting for nonlinearities and interactions, and the overall low amount of variance accounted for by trait predictors suggest limits to trait‐based statistical predictive frameworks. [ABSTRACT FROM AUTHOR]

Antarctic moss beds are sensitive to climatic conditions, and both their survival and community composition are particularly influenced by the availability of liquid water over summer. As Antarctic regions increasingly face climate pressures (e.g., changing hydrology and heat waves), advancing capabilities to efficiently and non‐destructively monitor water content in moss communities becomes a key research priority. Because of the complexity induced by multiple micro‐climatic drivers and its fragility, tracking the evolution and responses of moss bed moisture requires monitoring methods that are non‐intrusive, efficient, and spatially significant, such as the use of unoccupied aerial systems (UAS). In this study, we combine a multi‐species drying laboratory experiment with short‐wave infrared (SWIR) spectroscopy analyses to first develop a Random Forest regression Model (RFM) capable of predicting Antarctic moss turf water content (~5% error). The RFM was then applied to UAS‐borne SWIR imaging data (900–1700 nm, <16 nm spectral resolution) of the moss beds at high spatial resolution (2 cm) across three sites in the vicinity of Casey Station, Antarctica. The sites differed in terrain, snow cover, and moisture availability to evaluate method capabilities under different conditions. Optimum RFM parameters and input variables (spectral indices and reflectance spectra) were determined. Maps of moss moisture were validated via acquiring moss spectra and water content (using sponges inserted into the moss turf) collected in situ, for which an exponential correlation (R2 = 0.72) was reported. RFM further allowed investigation of the influential spectral variables to model water content in moss and associated spectral water absorption features. We demonstrated that UAS‐borne SWIR imaging is a promising new tool to map and quantify water content in Antarctic moss beds. Hyperspectral mapping facilitates the exploration of the spatial variability of moss health and enables the creation of a baseline against which changes in these moss communities can be measured. [ABSTRACT FROM AUTHOR]

A combined multiscale approach using ground, drone and satellite surveys can provide accurate landscape scale spatial mapping and monitoring. We used field observations with drone collected imagery covering 70 ha annually for a 5‐year period to estimate changes in living and dead vegetation of four widespread and abundant arid zone woody shrub species. Random forest classifiers delivered high accuracy (> 95%) using object‐based detection methods, with fast repeatable and transferrable processing using Google Earth Engine. Our classifiers performed well in both dominant arid zone landscape types: dune and swale, and at extremes of dry and wet years with minimal alterations. This highlighted the flexibility of the approach, potentially delivering insights into changes in highly variable environments. We also linked this classified drone vegetation to available temporally and spatially explicit Landsat satellite imagery, training a new, more accurate fractional vegetation cover model, allowing for accurate tracking of vegetation responses at large scales in the arid zone. Our method promises considerable opportunity to track vegetation dynamics including responses to management interventions, at large geographic scales, extending inference well beyond ground surveys. [ABSTRACT FROM AUTHOR]

Premise: To improve forest conservation monitoring, we developed a protocol to automatically count and identify the seeds of plant species with minimal resource requirements, making the process more efficient and less dependent on human operators. Methods and Results: Seeds from six North American conifer tree species were separated from leaf litter and imaged on a flatbed scanner. In the most successful species‐classification approach, an ImageJ macro automatically extracted measurements for random forest classification in the software R. The method allows for good classification accuracy, and the same process can be used to train the model on other species. Conclusions: This protocol is an adaptable tool for efficient and consistent identification of seed species or potentially other objects. Automated seed classification is efficient and inexpensive, making it a practical solution that enhances the feasibility of large‐scale monitoring projects in conservation biology. [ABSTRACT FROM AUTHOR]

Climate change majorly impacts biodiversity in diverse regions across the world, including South Asia, a megadiverse area with heterogeneous climatic and vegetation regions. However, climate impacts on bats in this region are not well‐studied, and it is unclear whether climate effects will follow patterns predicted in other regions. We address this by assessing projected near‐future changes in climatically suitable areas for 110 bat species from South Asia. We used ensemble ecological niche modelling with four algorithms (random forests, artificial neural networks, multivariate adaptive regression splines and maximum entropy) to define climatically suitable areas under current conditions (1970–2000). We then extrapolated near future (2041–2060) suitable areas under four projected scenarios (combining two global climate models and two shared socioeconomic pathways, SSP2: middle‐of‐the‐road and SSP5: fossil‐fuelled development). Projected future changes in suitable areas varied across species, with most species predicted to retain most of the current area or lose small amounts. When shifts occurred due to projected climate change, new areas were generally northward of current suitable areas. Suitability hotspots, defined as regions suitable for >30% of species, were generally predicted to become smaller and more fragmented. Overall, climate change in the near future may not lead to dramatic shifts in the distribution of bat species in South Asia, but local hotspots of biodiversity may be lost. Our results offer insight into climate change effects in less studied areas and can inform conservation planning, motivating reappraisals of conservation priorities and strategies for bats in South Asia. [ABSTRACT FROM AUTHOR]

Observing animals in the wild often poses extreme challenges, but animal‐borne accelerometers are increasingly revealing unobservable behaviours. Automated machine learning streamlines behaviour identification from the substantial datasets generated during multi‐animal, long‐term studies; however, the accuracy of such models depends on the qualities of the training data. We examined how data processing influenced the predictive accuracy of random forest (RF) models, leveraging the easily observed domestic cat (Felis catus) as a model organism for terrestrial mammalian behaviours. Nine indoor domestic cats were equipped with collar‐mounted tri‐axial accelerometers, and behaviours were recorded alongside video footage. From this calibrated data, eight datasets were derived with (i) additional descriptive variables, (ii) altered frequencies of acceleration data (40 Hz vs. a mean over 1 s) and (iii) standardised durations of different behaviours. These training datasets were used to generate RF models that were validated against calibrated cat behaviours before identifying the behaviours of five free‐ranging tag‐equipped cats. These predictions were compared to those identified manually to validate the accuracy of the RF models for free‐ranging animal behaviours. RF models accurately predicted the behaviours of indoor domestic cats (F‐measure up to 0.96) with discernible improvements observed with post‐data‐collection processing. Additional variables, standardised durations of behaviours and higher recording frequencies improved model accuracy. However, prediction accuracy varied with different behaviours, where high‐frequency models excelled in identifying fast‐paced behaviours (e.g. locomotion), whereas lower‐frequency models (1 Hz) more accurately identified slower, aperiodic behaviours such as grooming and feeding, particularly when examining free‐ranging cat behaviours. While RF modelling offered a robust means of behaviour identification from accelerometer data, field validations were important to validate model accuracy for free‐ranging individuals. Future studies may benefit from employing similar data processing methods that enhance RF behaviour identification accuracy, with extensive advantages for investigations into ecology, welfare and management of wild animals. [ABSTRACT FROM AUTHOR]

As the power source of the engine, the Fuel Pump(FP) plays a vital role in the safe operation of the aircraft. Due to the complexity of the working mechanism of Aviation Fuel Pumps (AFP) and the strong correlation between the components, the assessment model cannot reflect the health state of the FPs better, while the initial parameters in the assessment model will affect the assessment effect of the model. Therefore, this paper proposes a health status assessment model that can fully integrate monitoring data. To improve the accuracy of the model parameters, the Random Forest algorithm is used to give the feature weights to make up for the limitation of relying on expert knowledge, and the model parameters are optimized by the Covariance Matrix Adaptive Evolutionary Strategy algorithm, which achieves an accurate assessment of the state. Finally, the AFP test bed was built and the AFP was tested. Compared with other methods, the accuracy of the proposed method in this question reaches 96%, which is greatly superior to other methods and verifies the effectiveness of the proposed method. It also provides an outlook on future research directions for health state assessment. [ABSTRACT FROM AUTHOR]

Background: A lack of consensus exists in linking demographic, behavioral, and cognitive characteristics to biological stages of dementia, defined by the ATN (amyloid, tau, neurodegeneration) classification incorporating amyloid, tau, and neuronal injury biomarkers. Methods: Using a random forest classifier we investigated whether 27 demographic, behavioral, and cognitive characteristics allowed distinction between ATN‐defined groups with the same cognitive profile. This was done separately for three cognitively unimpaired (CU) (112 A‐T‐N‐; 46 A+T+N+/−; 65 A‐T+/‐N+/−) and three mild cognitive impairment (MCI) (128 A‐T‐N‐; 223 A+T+N+/−; 94 A‐T+/‐N+/−) subgroups. Results: Classification‐balanced accuracy reached 39% for the CU and 52% for the MCI subgroups. Logical Delayed Recall (explaining 16% of the variance), followed by the Alzheimer's Disease Assessment Scale 13 (14%) and Everyday Cognition Informant (10%), were the most relevant characteristics for classification of the MCI subgroups. Race and ethnicity, marital status, and Everyday Cognition Patient were not relevant (0%). Conclusions: The demographic, behavioral, and cognitive measures used in our model were not informative in differentiating ATN‐defined CU profiles. Measures of delayed memory, general cognition, and activities of daily living were the most informative in differentiating ATN‐defined MCI profiles; however, these measures alone were not sufficient to reach high classification performance. [ABSTRACT FROM AUTHOR]

Aim: In salt‐affected environments, salinity shapes ecosystem functions and species composition. Apart from salinity, however, we know little about how soil chemical factors affect plant species. We hypothesized that specific ions, most of which contribute to salinity, co‐determine plant niche differentiation. We asked if the importance of ions differs for species with (halophytes) and without (associated species) physiological adaptations to saline soils. Location: Carpatho‐Pannonian region (Central and Eastern Europe). Time period: 2005–2021. Major taxa studied: Vascular plants. Methods: We recorded species occurrences and collected soil samples in 433 plots in saline habitats. We measured pH, salinity (electrical conductivity), and concentrations of Ca2+, K+, Mg2+, Na+, SO42− Cl−, CO32− and mineral nitrogen (mN) and calculated the sodium adsorption ratio (SAR). For 88 species, we fitted response curves with Huisman–Olff–Fresco (HOF) models. To study ions' effects on species composition and ions' variance, we compared unconstrained and constrained ordinations and performed a principal component analysis. We used random forests to analyse the importance of ions for individual species and created two‐dimensional species niche plots for key ions. Results: Ion concentration niches varied among species and did not necessarily correspond to soil salinity or alkalinity. We frequently observed monotonic, sigmoidal model responses, while skewed unimodal responses were rare. Ions explained a considerable proportion of species compositional variation. Particularly, Na+, SO42−, Cl−, and CO32− contributed to the ions' variance. Na+, followed by SO42−, Cl−, CO32−, Ca2+, Mg2+, and mN, was most important for the occurrence of individual species. Compared to associated species, Na+, SO42−, and mN were significantly less important for halophytes, whereas Cl− and CO32− played a significant role. Main conclusions: We show that ion composition co‐determines niche differentiation in saline soils, suggesting evolved physiological adaptations in halophytes. Our study calls for incorporating high‐resolution data on soil ion composition in ecological research. [ABSTRACT FROM AUTHOR]

Soil moisture (SM) is a critical element of the hydrological cycle, land surface processes, and surface energy balance. However, the low spatial resolution of commonly used SM products limits the application of SM in agriculture and eco‐hydrology in cold and arid regions. In this study, the normalized difference soil index (NDSI) and bare soil index (BSI) were added to traditional downscaling factors including land surface temperature, normalized difference vegetation index, digital elevation mode, apparent thermal inertia, Albedo, and temperature vegetation dryness index, as they are more strongly correlated with surface SM in the bare soil‐vegetation alternation zone of such region. Using the random forest algorithm, a downscaling model of SM was constructed for such region. The accuracy of the downscaled SM estimates was validated by comparing them with the original SM data collected from May to September 2021, which is the non‐freezing period of the soil. The findings indicate that the newly added NDSI and BSI have good correlation with SM. Incorporating NDSI and BSI to construct the downscaled model enhances the accuracy by over 19% compared to excluding them, while also providing a more comprehensive representation of SM information. NDSI and BSI can be well applied to the downscaled research of SM in the bare soil‐vegetation alternation zone, which is of great value for the study of eco‐hydrology and agricultural drought monitoring in cold and arid regions. Core Ideas: Adding soil factors improves the downscaling of soil moisture in cold and arid regions.Compared with the traditional model, the improved model showed a significant increase in accuracy.This improved model better characterized soil moisture spatial and temporal variability. [ABSTRACT FROM AUTHOR]

Alzheimer's disease (AD) constitutes a neurodegenerative disorder marked by a progressive decline in cognitive function and memory capacity. The accurate diagnosis of this condition predominantly relies on cerebrospinal fluid (CSF) markers, notwithstanding the associated burdens of pain and substantial financial costs endured by patients. This study encompasses subjects exhibiting varying degrees of cognitive impairment, encompassing individuals with subjective cognitive decline, mild cognitive impairment, and dementia, constituting a total sample size of 82 participants. The primary objective of this investigation is to explore the relationships among brain atrophy measurements derived from magnetic resonance imaging, atypical electroencephalography (EEG) patterns, behavioral assessment scales, and amyloid β‐protein (Aβ) indicators. The findings of this research reveal that individuals displaying reduced Aβ1‐42/Aβ‐40 levels exhibit significant atrophy in the frontotemporal lobe, alongside irregularities in various parameters related to EEG frequency characteristics, signal complexity, inter‐regional information exchange, and microstates. The study additionally endeavors to estimate Aβ1‐42/Aβ‐40 content through the application of a random forest algorithm, amalgamating structural data, electrophysiological features, and clinical scales, achieving a remarkable predictive precision of 91.6%. In summary, this study proposes a cost‐effective methodology for acquiring CSF markers, thereby offering a valuable tool for the early detection of AD. [ABSTRACT FROM AUTHOR]

Among the most active oxygen evolution reaction (OER) catalysts, mixed metal oxides based on Ni, Fe, and Co metals are recognized as economical yet excellent replacements for RuO2 and IrOx. However, tuning and searching for optimal compositions of multi–element–compound electrocatalysts is a big challenge in catalysis research. Conventional materials screening experiments and theoretical simulations are labor–intensive and time–consuming. Machine learning offers a promising paradigm for accelerating electrocatalyst research and simultaneously understanding composition–activity correlation. Herein, we introduce an Explainable AI (XAI) framework for predicting the electrocatalytic performance of OER catalysts. By integrating the robust Random Forest (RF) model for machine learning with the Shapley Additive Explanations (SHAP) method for model explanation, we achieved accurate predictions of the overpotential for various compositions of (Ni−Fe−Co−Ce)Ox catalysts (R2=0.8221). More importantly, we obtained valuable insights into how each metal and their interactions influence the overpotential of the catalysts. Our results highlight the versatility of the RF model with SHAP in identifying the optimal composition of (Ni−Fe−Co−Ce)Ox catalysts for electrocatalytic oxygen evolution, showing its potential applicability across various catalyst synthesis methods. Finally, we anticipate that this work will lead to exciting possibilities in designing highly active multi–element compound electrocatalysts with the aid of explainable AI. [ABSTRACT FROM AUTHOR]

River algal blooms have become a global environmental problem due to their large impact range and environmental hazards. However, the complex mechanisms underlying these blooms make prediction and prevention challenging. Here, we employed empirical dynamic modeling (EDM) and machine learning to reveal the causes and predict diatom blooms from 2003 to 2017 in the Han River of China. The diatom cell density ranged from 0.1 to 5.1 × 107 cells L−1, whereas algal blooms often lasted for 10 days with density exceeding 107 cells L−1. The EDM results elucidated that, under consistent high nutrient concentrations, algal blooms were primarily regulated by eight environmental factors: water temperature in the Han River; water levels, flow velocities, and streamflow discharges in the Han River and the Yangtze River; and water level variation in the Han River. The poor performance (coefficient of determination R2 < 0) of the multiple linear regression, EDM, and random forest models indicated the challenge of predicting algae density. Therefore, we used machine learning classification models to predict algal blooms occurrences. With the resampling techniques to account for imbalanced data, machine learning models achieved perfect classification prediction (Kappa value = 1) of the 13 algae‐bloom events with a 10‐day lead time during a 15‐year period, providing an important reference for preemptive warning of riverine algal blooms. Plain Language Summary: River algal blooms have become a global environmental problem due to their large impact range and environmental hazards. The Han River is one of the most frequent areas of river algal blooms in China, but there is no clear conclusion on the causes of algal blooms formation. In this study, we combined two methods to reveal the causes and predict the occurrence of algal blooms in the Han River. Our results show that the prediction of algal blooms in the Han River is feasible. These blooms are primarily affected by the water temperature and seven other hydrological variables in the Han and Yangtze River. This study provides a starting point for the preemptive warning of riverine algal blooms. Key Points: Empirical dynamic modeling detects causality for river algal bloomsMachine learning accurately predicts the occurrence of algal blooms in the Han RiverResampling techniques improve robustness and accuracy of machine learning [ABSTRACT FROM AUTHOR]

Purposes: This study aimed to develop and validate algorithms for automating intensity modulated radiation therapy (IMRT) planning in breast cancer patients, with a focus on patient anatomical characteristics. Material and Methods: We retrospectively selected 400 breast cancer patients without lymph node involvement for automated treatment planning. Automation was achieved using the Eclipse Scripting Application Programming Interface (ESAPI) integrated into the Eclipse Treatment Planning System. We employed three beam insertion geometries and three optimization strategies, resulting in 3600 plans, each delivering a 40.05 Gy dose in 15 fractions. Gantry angles in the tangent fields were selected based on a criterion involving the minimum intersection area between the Planning Target Volume (PTV) and the ipsilateral lung in the Beam's Eye View projection. ESAPI was also used to gather patient anatomical data, serving as input for Random Forest models to select the optimal plan. The Random Forest classification considered both beam insertion geometry and optimization strategy. Dosimetric data were evaluated in accordance with the Radiation Therapy Oncology Group (RTOG) 1005 protocol. Results: Overall, all approaches generated high‐quality plans, with approximately 94% meeting the acceptable dose criteria for organs at risk and/or target coverage as defined by RTOG guidelines. Average automated plan generation time ranged from 6 min and 37 s to 9 min and 22 s, with the mean time increasing with additional fields. The Random Forest approach did not successfully enable automatic planning strategy selection. Instead, our automated planning system allows users to choose from the tested geometry and strategy options. Conclusions: Although our attempt to correlate patient anatomical features with planning strategy using machine learning tools was unsuccessful, the resulting dosimetric outcomes proved satisfactory. Our algorithm consistently produced high‐quality plans, offering significant time and efficiency advantages. [ABSTRACT FROM AUTHOR]

The fundamental goal of a rare plant translocation is to create self‐sustaining populations with the evolutionary resilience to persist in the long term. Yet, most plant translocation syntheses focus on a few factors influencing short‐term benchmarks of success (e.g., survival and reproduction). Short‐term benchmarks can be misleading when trying to infer future growth and viability because the factors that promote establishment may differ from those required for long‐term persistence. We assembled a large (n = 275) and broadly representative data set of well‐documented and monitored (7.9 years on average) at‐risk plant translocations to identify the most important site attributes, management techniques, and species' traits for six life‐cycle benchmarks and population metrics of translocation success. We used the random forest algorithm to quantify the relative importance of 29 predictor variables for each metric of success. Drivers of translocation outcomes varied across time frames and success metrics. Management techniques had the greatest relative influence on the attainment of life‐cycle benchmarks and short‐term population trends, whereas site attributes and species' traits were more important for population persistence and long‐term trends. Specifically, large founder sizes increased the potential for reproduction and recruitment into the next generation, whereas declining habitat quality and the outplanting of species with low seed production led to increased extinction risks and a reduction in potential reproductive output in the long‐term, respectively. We also detected novel interactions between some of the most important drivers, such as an increased probability of next‐generation recruitment in species with greater seed production rates, but only when coupled with large founder sizes. Because most significant barriers to plant translocation success can be overcome by improving techniques or resolving site‐level issues through early intervention and management, we suggest that by combining long‐term monitoring with adaptive management, translocation programs can enhance the prospects of achieving long‐term success. [ABSTRACT FROM AUTHOR]

Ensemble of label ranking trees (LRTs) are currently the state‐of‐the‐art approaches to the label ranking problem. Recently, bagging, boosting, and random forest methods have been proposed, all based on the LRT algorithm, which adapts regression/classification trees to the label classification problem. The LRT algorithm uses theoretically grounded Mallows probability distribution to select the best split when growing the tree, and an EM‐type process to complete the rankings on the training data when they are incomplete. These two steps have proven to be accurate, but require a large computational effort. This article proposes two alternative methods that replace the use of the Mallows distribution with distance‐based criteria to select the best split at each inner node of the tree. Moreover, these distance‐based criteria allow dealing with incomplete rankings natively, so avoiding the completion process. We have carried out an extensive experimental evaluation, which shows that (1) the integration of the two proposed modifications to the LRT algorithm into ensemble methods (bagging and random forest) are an order of magnitude faster than using the original Mallows‐based LRT algorithm; (2) ensembles using the proposed LRT methods are significantly more accurate in the presence of incomplete rankings, while they are at least as accurate in the complete case; and (3) the two modified LRT algorithms are also an order of magnitude faster than the Mallows‐based LRT, while they are at least as accurate as the Mallows‐based LRT on both complete and incomplete rankings. [ABSTRACT FROM AUTHOR]