Search

To accurately and rapidly predict seismic responses, including the maximum displacement (MaxD) and maximum acceleration (MaxA), of the isolated structure considering the soil–structure interaction (SSI), five ensemble learning models, i.e. random forest (RF), gradient boosting regression tree (GBRT), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM) and stacking model, are constructed. Firstly, a total of 96 000 nonlinear time history analyses of the isolated structure considering the SSI are conducted with the aid of OpenSees. The generated database is used for training and testing ensemble learning models. The ensemble learning models have 12 input variables in four categories, i.e. ground motion parameters, structural parameter, isolation parameters and soil parameter, and two output variables, i.e. MaxD and MaxA. The study shows that all ensemble learning models have excellent prediction performance for both training and testing datasets. The determination coefficients are larger than 0.96 and root-mean-square errors (RMSEs) are relatively small. Among the five ensemble learning models, the stacking model exhibits the best performance. In addition, the calculation method of feature importance score for the stacking model is provided. According to the feature importance analysis, the ground motion parameters have greater impact on seismic responses than other three categories of inputs. Finally, six ground motions are randomly selected to verify the generalization ability of the proposed ensemble learning models. The results show that the stacking model has a favorable generalization ability with relatively small prediction errors. [ABSTRACT FROM AUTHOR]

After adopting a combined approach of data-driven methods and machine learning, the prediction of material performance and the optimization of composition design can significantly reduce the development time of materials at a lower cost. In this research, we employed four machine learning algorithms, including linear regression, ridge regression, support vector regression, and backpropagation neural networks, to develop predictive models for the electrical performance data of titanium alloys. Our focus was on two key objectives: resistivity and the temperature coefficient of resistance (TCR). Subsequently, leveraging the results of feature selection, we conducted an analysis to discern the impact of alloying elements on these two electrical properties.The prediction results indicate that for the resistivity data prediction task, the radial basis function kernel-based support vector machine model performs the best, with a correlation coefficient above 0.995 and a percentage error within 2%, demonstrating high predictive capability. For the TCR data prediction task, the best-performing model is a backpropagation neural network with two hidden layers, also with a correlation coefficient above 0.995 and a percentage error within 3%, demonstrating good generalization ability. The feature selection results using random forest and Xgboost indicate that Al and Zr have a significant positive effect on resistivity, while Al, Zr, and V have a significant negative effect on TCR. The conclusion of the composition optimization design suggests that to achieve both high resistivity and TCR, it is recommended to set the Al content in the range of 1.5% to 2% and the Zr content in the range of 2.5% to 3%. [ABSTRACT FROM AUTHOR]

Mass-spectrometry-based non-targeted analysis (NTA), in which mass spectrometric signals are assigned chemical identities based on a systematic collation of evidence, is a growing area of interest for toxicological risk assessment. Successful NTA results in better identification of potentially hazardous pollutants within the environment, facilitating the development of targeted analytical strategies to best characterize risks to human and ecological health. A supporting component of the NTA process involves assessing whether suspected chemicals are amenable to the mass spectrometric method, which is necessary in order to assign an observed signal to the chemical structure. Prior work from this group involved the development of a random forest model for predicting the amenability of 5517 unique chemical structures to liquid chromatography–mass spectrometry (LC-MS). This work improves the interpretability of the group's prior model of the same endpoint, as well as integrating 1348 more data points across negative and positive ionization modes. We enhance interpretability by feature engineering, a machine learning practice that reduces the input dimensionality while attempting to preserve performance statistics. We emphasize the importance of interpretable machine learning models within the context of building confidence in NTA identification. The novel data were curated by the labeling of compounds as amenable or unamenable by expert curators, resulting in an enhanced set of chemical compounds to expand the applicability domain of the prior model. The balanced accuracy benchmark of the newly developed model is comparable to performance previously reported (mean CV BA is 0.84 vs. 0.82 in positive mode, and 0.85 vs. 0.82 in negative mode), while on a novel external set, derived from this work's data, the Matthews correlation coefficients (MCC) for the novel models are 0.66 and 0.68 for positive and negative mode, respectively. Our group's prior published models scored MCC of 0.55 and 0.54 on the same external sets. This demonstrates appreciable improvement over the chemical space captured by the expanded dataset. This work forms part of our ongoing efforts to develop models with higher interpretability and higher performance to support NTA efforts. [ABSTRACT FROM AUTHOR]

Background: Oral microbiota imbalance is associated with the progression of various lung diseases, including lung cancer. Pulmonary nodules (PNs) are often considered a critical stage for the early detection of lung cancer; however, the relationship between oral microbiota and PNs remains unknown. Methods: We conducted a 'Microbiome with pulmonary nodule series study 1' (MCEPN-1) where we compared PN patients and healthy controls (HCs), aiming to identify differences in oral microbiota characteristics and discover potential microbiota biomarkers for non-invasive, radiation-free PNs diagnosis and warning in the future. We performed 16 S rRNA amplicon sequencing on saliva samples from 173 PN patients and 40 HCs to compare the characteristics and functional changes in oral microbiota between the two groups. The random forest algorithm was used to identify PN salivary microbial markers. Biological functions and potential mechanisms of differential genes in saliva samples were preliminarily explored using the Kyoto Encyclopedia of Genes and Genomes (KEGG) and Cluster of Orthologous Groups (COG) analyses. Results: The diversity of salivary microorganisms was higher in the PN group than in the HC group. Significant differences were noted in community composition and abundance of oral microorganisms between the two groups. Neisseria, Prevotella, Haemophilus and Actinomyces, Porphyromonas, Fusobacterium, 7M7x, Granulicatella and Selenomonas were the main differential genera between the PN and HC groups. Fusobacterium, Porphyromonas, Parvimonas, Peptostreptococcus and Haemophilus constituted the optimal marker sets (area under curve, AUC = 0.80), which can distinguish between patients with PNs and HCs. Further, the salivary microbiota composition was significantly correlated with age, sex, and smoking history (P < 0.001), but not with personal history of cancer (P > 0.05). Bioinformatics analysis of differential genes showed that patients with PN showed significant enrichment in protein/molecular functions related to immune deficiency and energy metabolisms, such as the cytoskeleton protein RodZ, nicotinamide adenine dinucleotide phosphate dehydrogenase (NADPH) dehydrogenase, major facilitator superfamily transporters and AraC family transcription regulators. Conclusions: Our study provides the first evidence that the salivary microbiota can serve as potential biomarkers for identifying PN. We observed a significant association between changes in the oral microbiota and PNs, indicating the potential of salivary microbiota as a new non-invasive biomarker for PNs. Trial registration: Clinical trial registration number: ChiCTR2200062140; Date of registration: 07/25/2022. [ABSTRACT FROM AUTHOR]

The estimation of material performance plays a crucial role in practical life, enabling the rational allocation of time and resources while enhancing the practical application of materials. Therefore, this study investigates the analytical and predictive capabilities of five machine learning (ML) algorithms (the Random Forest algorithm – RF, the Adaboost algorithm – AB, the Gradient Boosting algorithm – GB, the Extra Trees algorithm – ET and linear models – LM) for comprehensive performance parameters of wood ceramics (ammonia nitrogen adsorption quantity - Q , open porosity - P , burn-off , and hardness). In the analysis of model prediction capabilities, five key statistical parameters (Root Mean Square Error - RMSE , Mean Square Error - MSE , Coefficient of Correlation - R , Determination Coefficient - R 2 , and Mean Absolute Relative Error - MARE) were calculated. The results indicate the following: (1) Among various ML models, the GB model exhibits the most superior performance with an R 2 ≥ 0.9594. (2) However, in predicting the four performance parameters of wood ceramics, the AB model with an R 2 ranging from 0.0078 to 0.45, indicating notably poor predictive capabilities. Despite integrating the Support Vector Regression (SVR) module, no enhancement in predictive accuracy was observed. (3) While forecasting the four performance parameters of wood ceramics, the LM model demonstrates predictive accuracy relatively akin to that of the RF and ET models among the ML models. (4) Though the feature importance scores vary across distinct input variable models, there is a consistent trend of change. (5) Variables B and D exhibit some level of correlation with other variables. In summary, the results of this study suggest that, for the analysis and prediction of wood ceramic performance, the GB model among the five regression models demonstrates outstanding simulation performance. [ABSTRACT FROM AUTHOR]

In real-world classification problems, it is important to build accurate prediction models and provide information that can improve decision-making. Decision-support tools are often based on network models, and this article uses information encoded by social networks to solve the problem of employer turnover. However, understanding the factors behind black-box prediction models can be challenging. Our question was about the predictability of employee turnover, given information from the multilayer network that describes collaborations and perceptions that assess the performance of organizations that indicate the success of cooperation. Our goal was to develop an accurate prediction procedure, preserve the interpretability of the classification, and capture the wide variety of specific reasons that explain positive cases. After a feature engineering, we identified variables with the best predictive power using decision trees and ranked them based on their added value considering their frequent co-occurrence. We applied the Random Forest using the SMOTE balancing technique for prediction. We calculated the SHAP values to identify the variables that contribute the most to individual predictions. As a last step, we clustered the sample based on SHAP values to fine-tune the explanations for quitting due to different background factors. [ABSTRACT FROM AUTHOR]

Estimating penetration rates of Jumbo drills is crucial for optimizing underground mining drilling processes, aiming to reduce costs and time. This study investigates various regression and machine learning methods, including Multilayer Perceptron (MLP), Support Vector Regression (SVR), and Random Forests (RF), to predict the penetration rates (ROP) using multivariate inputs such as operation parameters and rock mass characteristics. The Rock Mass Drillability Index (RDi), incorporating both intact rock properties and structural parameters, was utilized to characterize the rock mass. The dataset was split into 80% for training and 20% for testing. Performance metrics including correlation coefficient (R2), variance accounted for (VAF), mean absolute error (MAE), mean absolute percentage error (MAPE), and root mean square error (RMSE) were calculated for each method to evaluate the accuracy of the predictions. SVR exhibited the best prediction performance for ROP, achieving the highest R2, lowest RMSE, MAE, and MAPE, as well as the largest VAF values of 0.94, 0.15, 0.11, 4.84, and 94.13 during training, and 0.91, 0.19, 0.13, 6.02, and 91.11 during testing, respectively. With this high accuracy, we conclude that the proposed machine learning algorithms are valuable and efficient predictors for estimating jumbo drill penetration rates in underground mining operations. [ABSTRACT FROM AUTHOR]

Ischemic stroke (IS) is a common cerebrovascular disease whose pathogenesis involves a variety of immune molecules, immune channels and immune processes. 6-methyladenosine (m6A) modification regulates a variety of immune metabolic and immunopathological processes, but the role of m6A in IS is not yet understood. We downloaded the data set GSE58294 from the GEO database and screened for m6A-regulated differential expression genes. The RF algorithm was selected to screen the m6A key regulatory genes. Clinical prediction models were constructed and validated based on m6A key regulatory genes. IS patients were grouped according to the expression of m6A key regulatory genes, and immune markers of IS were identified based on immune infiltration characteristics and correlation. Finally, we performed functional enrichment, protein interaction network analysis and molecular prediction of the immune biomarkers. We identified a total of 7 differentially expressed genes in the dataset, namely METTL3, WTAP, YWHAG, TRA2A, YTHDF3, LRPPRC and HNRNPA2B1. The random forest algorithm indicated that all 7 genes were m6A key regulatory genes of IS, and the credibility of the above key regulatory genes was verified by constructing a clinical prediction model. Based on the expression of key regulatory genes, we divided IS patients into 2 groups. Based on the expression of the gene LRPPRC and the correlation of immune infiltration under different subgroups, LRPPRC was identified as an immune biomarker for IS. GO enrichment analyses indicate that LRPPRC is associated with a variety of cellular functions. Protein interaction network analysis and molecular prediction indicated that LRPPRC correlates with a variety of immune proteins, and LRPPRC may serve as a target for IS drug therapy. Our findings suggest that LRPPRC is an immune marker for IS. Further analysis based on LRPPRC could elucidate its role in the immune microenvironment of IS. [ABSTRACT FROM AUTHOR]

Many individual studies in the literature observed the superiority of tree-based machine learning (ML) algorithms. However, the current body of literature lacks statistical validation of this superiority. This study addresses this gap by employing five ML algorithms on 200 open-access datasets from a wide range of research contexts to statistically confirm the superiority of tree-based ML algorithms over their counterparts. Specifically, it examines two tree-based ML (Decision tree and Random forest) and three non-tree-based ML (Support vector machine, Logistic regression and k-nearest neighbour) algorithms. Results from paired-sample t-tests show that both tree-based ML algorithms reveal better performance than each non-tree-based ML algorithm for the four ML performance measures (accuracy, precision, recall and F1 score) considered in this study, each at p<0.001 significance level. This performance superiority is consistent across both the model development and test phases. This study also used paired-sample t-tests for the subsets of the research datasets from disease prediction (66) and university-ranking (50) research contexts for further validation. The observed superiority of the tree-based ML algorithms remains valid for these subsets. Tree-based ML algorithms significantly outperformed non-tree-based algorithms for these two research contexts for all four performance measures. We discuss the research implications of these findings in detail in this article. [ABSTRACT FROM AUTHOR]

Purpose: To explore the feasibility and validity of machine learning models in determining causality in medical malpractice cases and to try to increase the scientificity and reliability of identification opinions. Methods: We collected 13,245 written judgments from PKULAW.COM, a public database. 963 cases were included after the initial screening. 21 medical and ten patient factors were selected as characteristic variables by summarising previous literature and cases. Random Forest, eXtreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM) were used to establish prediction models of causality for the two data sets, respectively. Finally, the optimal model is obtained by hyperparameter tuning of the six models. Results: We built three real data set models and three virtual data set models by three algorithms, and their confusion matrices differed. XGBoost performed best in the real data set, with a model accuracy of 66%. In the virtual data set, the performance of XGBoost and LightGBM was basically the same, and the model accuracy rate was 80%. The overall accuracy of external verification was 72.7%. Conclusions: The optimal model of this study is expected to predict the causality accurately. [ABSTRACT FROM AUTHOR]

Objectives: Carpal tunnel syndrome (CTS) stands as the most prevalent upper extremity entrapment neuropathy, with a multifaceted etiology encompassing various risk factors. This study aimed to investigate whether anthropometric measurements of the hand, grip strength, and pinch strength could serve as predictive indicators for CTS through machine learning techniques. Methods: Enrollment encompassed patients exhibiting CTS symptoms (n = 56) and asymptomatic healthy controls (n = 56), with confirmation via electrophysiological assessments. Anthropometric measurements of the hand were obtained using a digital caliper, grip strength was gauged via a digital handgrip dynamometer, and pinch strengths were assessed using a pinchmeter. A comprehensive analysis was conducted employing four most common and effective machine learning algorithms, integrating thorough parameter tuning and cross-validation procedures. Additionally, the outcomes of variable importance were presented. Results: Among the diverse algorithms, Random Forests (accuracy of 89.474%, F1-score of 0.905, and kappa value of 0.789) and XGBoost (accuracy of 86.842%, F1-score of 0.878, and kappa value of 0.736) emerged as the top-performing choices based on distinct classification metrics. In addition, using variable importance calculations specific to these models, the most important variables were found to be wrist circumference, hand width, hand grip strength, tip pinch, key pinch, and middle finger length. Conclusion: The findings of this study demonstrated that wrist circumference, hand width, hand grip strength, tip pinch, key pinch, and middle finger length can be utilized as reliable indicators of CTS. Also, the model developed herein, along with the identified crucial variables, could serve as an informative guide for healthcare professionals, enhancing precision and efficacy in CTS prediction. [ABSTRACT FROM AUTHOR]

The work presents a tool enabling the selection of powder for 3D printing. The project focused on three types of powders, such as steel, nickel- and cobalt-based and aluminum-based. An important aspect during the research was the possibility of obtaining the mechanical parameters. During the work, the possibility of using the selected algorithm based on artificial intelligence like Random Forest, Decision Tree, K-Nearest Neighbors, Fuzzy K-Nearest Neighbors, Gradient Boosting, XGBoost, AdaBoost was also checked. During the work, tests were carried out to check which algorithm would be best for use in the decision support system being developed. Cross-validation was used, as well as hyperparameter tuning using different evaluation sets. In both cases, the best model turned out to be Random Forest, whose F1 metric score is 98.66% for cross-validation and 99.10% after tuning on the test set. This model can be considered the most promising in solving this problem. The first result is a more accurate estimate of how the model will behave for new data, while the second model talks about possible improvement after optimization or possible overtraining to the parameters. [ABSTRACT FROM AUTHOR]

In order to realize the accurate and reliable fault diagnosis of hydraulic systems, a diagnostic model based on improved tuna swarm optimization (ITSO), optimized convolutional neural networks (CNNs), and bi-directional long short-term memory (BiLSTM) networks is proposed. Firstly, sensor selection is implemented using the random forest algorithm to select useful signals from six kinds of physical or virtual sensors including pressure, temperature, flow rate, vibration, motor power, and motor efficiency coefficient. After that, fused features are extracted by CNN, and then, BiLSTM is applied to learn the forward and backward information contained in the data. The ITSO algorithm is adopted to adaptively optimize the learning rate, regularization coefficient, and node number to obtain the optimal CNN-BiLSTM network. Improved Chebyshev chaotic mapping and the nonlinear reduction strategy are adopted to improve population initialization and individual position updating, further promoting the optimization effect of TSO. The experimental results show that the proposed method can automatically extract fusion features and effectively utilize multi-sensor information. The diagnostic accuracies of the plunger pump, cooler, throttle valve, and accumulator are 99.07%, 99.4%, 98.81%, and 98.51%, respectively. The diagnostic results of noisy data with 0 dB, 5 dB, and 10 dB signal-to-noise ratios (SNRs) show that the ITSO-CNN-BiLSTM model has good robustness to noise interference. [ABSTRACT FROM AUTHOR]

Geological hazards in Xinxian County, Xinyang City, Henan Province, are characterized by their small scale, wide distribution, and significant influence from regional tectonics. This study focuses on collapses and landslide hazards within the area, selecting twelve evaluation factors: aspect, slope shape, normalized difference vegetation index (NDVI), topographic relief, distance from geological structure, slope, distance from roads, land use cover type, area of land change (2012–2022), average annual rainfall (2012–2022), and river network density. Utilizing data from historical disaster sites across the region, the information quantity method and hierarchical analysis method are employed to ascertain the information quantity and weight of each factor. Subsequently, a random forest model is applied to perform susceptibility zoning of geological hazards in Xinxian County and to examine the characteristics of these geological disasters. The results show that in the study area, the primary factors influencing the development of geohazards are the distance from roads, rock groups, and distance from geological structure areas. A comparison of the susceptibility results obtained through two methods, the analytic hierarchy process information quantity method and the random forests model, reveals that the former exhibits a higher accuracy. This model categorizes the geohazard susceptibility in the study area into four levels: low, medium, high, and very high. Notably, the areas of very high and high susceptibility together cover 559.17 km2, constituting 35.99% of the study area's total area, and encompass 57 disaster sites, which represent 72.15% of all disaster sites. Geological hazards in Xinxian County frequently manifest on steep canyon inclines, along the curved and concave banks of mountain rivers, within watershed regions, on gully inclines, atop steep cliffs, and on artificially created slopes, among other sites. Areas with very high and high vulnerability to these hazards are mainly concentrated near the county's geological formations. The gneiss formations are widely exposed in Xinxian County, and the gneisses' strength is significantly changed under weathering, which makes the properties of the different degrees of weathering of the rock and soil bodies play a decisive role in the stability of the slopes. This paper provides a basis for evaluating and preventing geologic hazards in the Dabie mountainous area of the South Henan Province, and the spatial planning of the national territory. [ABSTRACT FROM AUTHOR]

Arid zone grassland is a crucial component of terrestrial ecosystems and plays a significant role in ecosystem protection and soil erosion prevention. However, accurately mapping grassland spatial information in arid zones presents a great challenge. The accuracy of remote sensing grassland mapping in arid zones is affected by spectral variability caused by the highly diverse landscapes. In this study, we explored the potential of a rectangular tile classification model, constructed using the random forest algorithm and integrated images from Sentinel-1A (synthetic aperture radar imagery) and Sentinel-2 (optical imagery), to enhance the accuracy of grassland mapping in the semiarid to arid regions of Ordos, China. Monthly Sentinel-1A median value images were synthesised, and four MODIS vegetation index mean value curves (NDVI, MSAVI, NDWI and NDBI) were used to determine the optimal synthesis time window for Sentinel-2 images. Seven experimental groups, including 14 experimental schemes based on the rectangular tile classification model and the traditional global classification model, were designed. By applying the rectangular tile classification model and Sentinel-integrated images, we successfully identified and extracted grasslands. The results showed the integration of vegetation index features and texture features improved the accuracy of grassland mapping. The overall accuracy of the Sentinel-integrated images from EXP7-2 was 88.23%, which was higher than the accuracy of the single sensor Sentinel-1A (53.52%) in EXP2-2 and Sentinel-2 (86.53%) in EXP5-2. In all seven experimental groups, the rectangular tile classification model was found to improve overall accuracy (OA) by 1.20% to 13.99% compared to the traditional global classification model. This paper presents novel perspectives and guidance for improving the accuracy of remote sensing mapping for land cover classification in arid zones with highly diverse landscapes. The study presents a flexible and scalable model within the Google Earth Engine framework, which can be readily customized and implemented in various geographical locations and time periods. [ABSTRACT FROM AUTHOR]

Detecting epistatic drivers of human phenotypes is a considerable challenge. Traditional approaches use regression to sequentially test multiplicative interaction terms involving pairs of genetic variants. For higher-order interactions and genome-wide large-scale data, this strategy is computationally intractable. Moreover, multiplicative terms used in regression modeling may not capture the form of biological interactions. Building on the Predictability, Computability, Stability (PCS) framework, we introduce the epiTree pipeline to extract higher-order interactions from genomic data using tree-based models. The epiTree pipeline first selects a set of variants derived from tissue-specific estimates of gene expression. Next, it uses iterative random forests (iRF) to search training data for candidate Boolean interactions (pairwise and higher-order). We derive significance tests for interactions, based on a stabilized likelihood ratio test, by simulating Boolean tree-structured null (no epistasis) and alternative (epistasis) distributions on hold-out test data. Finally, our pipeline computes PCS epistasis p-values that probabilisticly quantify improvement in prediction accuracy via bootstrap sampling on the test set. We validate the epiTree pipeline in two case studies using data from the UK Biobank: predicting red hair and multiple sclerosis (MS). In the case of predicting red hair, epiTree recovers known epistatic interactions surrounding MC1R and novel interactions, representing non-linearities not captured by logistic regression models. In the case of predicting MS, a more complex phenotype than red hair, epiTree rankings prioritize novel interactions surrounding HLA-DRB1, a variant previously associated with MS in several populations. Taken together, these results highlight the potential for epiTree rankings to help reduce the design space for follow up experiments. [ABSTRACT FROM AUTHOR]

Background: Severe burns may alter the stability of the intestinal flora and affect the patient's recovery process. Understanding the characteristics of the gut microbiota in the acute phase of burns and their association with phenotype can help to accurately assess the progression of the disease and identify potential microbiota markers. Methods: We established mouse models of partial thickness deep III degree burns and collected faecal samples for 16 S rRNA amplification and high throughput sequencing at two time points in the acute phase for independent bioinformatic analysis. Results: We analysed the sequencing results using alpha diversity, beta diversity and machine learning methods. At both time points, 4 and 6 h after burning, the Firmicutes phylum content decreased and the content of the Bacteroidetes phylum content increased, showing a significant decrease in the Firmicutes/Bacteroidetes ratio compared to the control group. Nine bacterial genera changed significantly during the acute phase and occupied the top six positions in the Random Forest significance ranking. Clustering results also clearly showed that there was a clear boundary between the communities of burned and control mice. Functional analyses showed that during the acute phase of burn, gut bacteria increased lipoic acid metabolism, seleno-compound metabolism, TCA cycling, and carbon fixation, while decreasing galactose metabolism and triglyceride metabolism. Based on the abundance characteristics of the six significantly different bacterial genera, both the XGboost and Random Forest models were able to discriminate between the burn and control groups with 100% accuracy, while both the Random Forest and Support Vector Machine models were able to classify samples from the 4-hour and 6-hour burn groups with 86.7% accuracy. Conclusions: Our study shows an increase in gut microbiota diversity in the acute phase of deep burn injury, rather than a decrease as is commonly believed. Severe burns result in a severe imbalance of the gut flora, with a decrease in probiotics and an increase in microorganisms that trigger inflammation and cognitive deficits, and multiple pathways of metabolism and substance synthesis are affected. Simple machine learning model testing suggests several bacterial genera as potential biomarkers of severe burn phenotypes. [ABSTRACT FROM AUTHOR]

Short-term load forecasting (STLF) plays a vital role in ensuring the safe, efficient, and economical operation of power systems. Accurate load forecasting provides numerous benefits for power suppliers, such as cost reduction, increased reliability, and informed decision-making. However, STLF is a complex task due to various factors, including non-linear trends, multiple seasonality, variable variance, and significant random interruptions in electricity demand time series. To address these challenges, advanced techniques and models are required. This study focuses on the development of an efficient short-term power load forecasting model using the random forest (RF) algorithm. RF combines regression trees through bagging and random subspace techniques to improve prediction accuracy and reduce model variability. The algorithm constructs a forest of trees using bootstrap samples and selects random feature subsets at each node to enhance diversity. Hyperparameters such as the number of trees, minimum sample leaf size, and maximum features for each split are tuned to optimize forecasting results. The proposed model was tested using historical hourly load data from four transformer substations supplying different campus areas of the University of Beira Interior, Portugal. The training data were from January 2018 to December 2021, while the data from 2022 were used for testing. The results demonstrate the effectiveness of the RF model in forecasting short-term hourly and one day ahead load and its potential to enhance decision-making processes in smart grid operations. [ABSTRACT FROM AUTHOR]

Accurate well productivity prediction plays a significant role in formulating reservoir development plans. However, traditional well productivity prediction methods lack accuracy in tight gas reservoirs; therefore, this paper quantitatively evaluates the correlations between absolute open flow and the critical parameters for Linxing tight gas reservoirs through statistical analysis. Dominant control factors are obtained by considering reservoir engineering theories, and a novel machine learning-based well productivity prediction method is proposed for tight gas reservoirs. The adaptability of the productivity prediction model is assessed through machine learning and field data analysis. Combined with the typical decline curve analysis, the estimated ultimate recovery (EUR) of a single well in the tight gas reservoir is forecasted in an appropriate range. The results of the study include 10 parameters (such as gas saturation) identified as the dominant controlling factors for well productivity and geological factors that impact the productivity in this area compared to fracturing parameters. According to the prediction results of the three models, the R2 of Support Vector Regression (SVR), Back Propagation (BP), and Random Forest (RF) models are 0.72, 0.87, and 0.91, respectively. The results indicate that RF has a more accurate prediction. In addition, the RF model is more suitable for medium and high-production wells based on the actual field data. Based on this model, it is verified that the productivity of low-producing wells is affected by water production. This study confirms the model's reliability and application value by predicting recoverable reserves for a single well. [ABSTRACT FROM AUTHOR]

Red needle cast (RNC), mainly caused by Phytophthora pluvialis, is a very damaging disease of the widely grown species radiata pine within New Zealand. Using a combination of satellite imagery and weather data, a novel methodology was developed to pre-visually predict the incidence of RNC on radiata pine within the Gisborne region of New Zealand over a five-year period from 2019 to 2023. Sentinel-2 satellite imagery was used to classify areas within the region as being disease-free or showing RNC expression from the difference in the red/green index (R/Gdiff) during a disease-free time of the year and the time of maximum disease expression in the upper canopy (early spring–September). Within these two classes, 1976 plots were extracted, and a classification model was used to predict disease incidence from mean monthly weather data for key variables during the 11 months prior to disease expression. The variables in the final random forest model included solar radiation, relative humidity, rainfall, and the maximum air temperature recorded during mid–late summer, which provided a pre-visual prediction of the disease 7–8 months before its peak expression. Using a hold-out test dataset, the final random forest model had an accuracy of 89% and an F1 score of 0.89. This approach can be used to mitigate the impact of RNC by focusing on early surveillance and treatment measures. [ABSTRACT FROM AUTHOR]

Accurate mapping of forest habitats, especially in NATURA sites, is essential information for forest monitoring and sustainable management but also for habitat characterisation and ecosystem functioning. Remote sensing data and spatial modelling allow accurate mapping of the presence and distribution of tree species and habitats and are valuable tools for the long-term assessment of habitat status required by the European Commission. In order to serve the above, the present study aims to propose a methodology to accurately map the spatial distribution of forest habitats in three NATURA2000 sites of Cyprus by employing Sentinel-1 and Sentinel-2 data as well as topographic features using the Google Earth Engine (GEE). A pivotal aspect of the methodology identified was that the best band combination of the Random Forest (RF) classifier achieves the highest performance for mapping the dominant habitats in the three case studies. Specifically, in the Akamas region, eight habitat types have been mapped, in Paphos nine and six in Troodos. These habitat types are included in three of the nine habitat groups based on the EU's Habitat Directive: the sclerophyllous scrub, rocky habitats and caves and forests. The results show that using the RF algorithm achieves the highest performance, especially using Dataset 6, which is based on S2 bands, spectral indices and topographical features, and Dataset 13, which includes S2, S1, spectral indices and topographical features. These datasets achieve an overall accuracy (OA) of approximately 91–94%. In contrast, Dataset 7, which includes only S1 bands and Dataset 9, which combines S1 bands and spectral indices, achieve the lowest performance with an OA of approximately 25–43%. [ABSTRACT FROM AUTHOR]

Maritime Antarctica (M.A.) contains the most extensive and diverse lithological exposure compared to the entire continent. This lithological substrate reveals a rich history encompassing lithological, pedogeomorphological, and glaciological aspects of M.A., all influenced by periglacial processes. Although geophysical surveys can detect and provide valuable information to understand Antarctic lithologies and their history, such surveys are scarce on this continent and, in practice, almost non-existent. In this sense, we conducted a pioneering and comprehensive gamma-spectrometric (natural radioactivity) and magnetic susceptibility (κ) survey on various igneous rocks. The main objective was to create ternary gamma-ray and κ maps using machine learning algorithms, terrain attributes, and a nested-leave-one-out cross-validation method. Additionally, we investigated the relationship between the distribution of natural radioactivity and κ to gain insights into pedogeomorphological and periglacial processes and dynamics. For that, we used proximal gamma-spectrometric and κ data in different lithological substrates associated to terrain attributes. The geophysical variables were collected in the field from various lithological substrates, by use field portable equipment. The geophysical variables were collected in the field from various lithological substrates using portable equipment. These variables, combined with relief data and lithology, served as input data for modeling to predict and spatially map the content of radionuclides and κ by random forest algorithm (RF). In addition, we use nested-LOOCV as a form of external validation in a geophysical data with a small number of samples, and the error maps as evaluation of results. The RF algorithm successfully generated detailed maps of gamma-spectrometric and κ variables. The distribution of radionuclides and ferrimagnetic minerals was influenced by morphometric variables. Nested-LOOCV method evaluated algorithm performance accurately with limited samples, generating robust mean maps. The highest thorium levels were observed in elevated, flat, and west beach areas, where detrital materials from periglacial erosion came through fluvioglacial channels. Lithology and pedogeomorphological processes-controlled thorium contents. Steeper areas formed a ring with the highest uranium contents, influenced by lithology and geomorphological-periglacial processes (rock cryoclasty, periglacial erosion, and heterogeneous deposition). Felsic rocks and areas less affected by periglacial erosion had the highest potassium levels, while regions with sulfurization-affected pyritized-andesites near fluvioglacial channels showed the lowest potassium contents. Lithology and pedogeochemical processes governed potassium levels. The κ values showed no distinct distribution pattern. Pyritized-andesite areas had the highest levels due to sulfurization and associated pyrrhotite, promoting iron release. Conversely, Cryosol areas, experiencing freezing and thawing activity, had the lowest κ values due to limited ferrimagnetic mineral formation. Lithology and pedological-periglacial processes in Cryosols played a significant role in controlling κ values. In regions characterized by diverse terrain attributes and abundant active and intense periglacial processes, the spatial distribution of geophysical variables does not reliably reflect the actual lithological composition of the substrate. The complex interplay of various periglacial processes in the area, along with the morphometric features of the landscape, leads to the redistribution, mixing, and homogenization of surface materials, contributing to the inaccuracies in the predicted-spatialized geophysical variables. [ABSTRACT FROM AUTHOR]

While the available set of gamma-ray burst (GRB) data with known redshift is currently limited, a much larger set of GRB data without redshift is available from different instruments. This data includes well-measured prompt gamma-ray flux and spectral information. We estimate the redshift of a selection of these GRBs detected by Fermi -GBM and Konus- Wind using machine learning techniques that are based on spectral parameters. We find that Deep Neural Networks with Random Forest models employing non-linear relations among input parameters can reasonably reproduce the pseudo-redshift distribution of GRBs, mimicking the distribution of GRBs with spectroscopic redshift. Furthermore, we find that the pseudo-redshift samples of GRBs to satisfy (i) Amati relation between the peak photon energy of the time-averaged energy spectrum in the cosmological rest frame of the GRB E i,p and the isotropic-equivalent radiated energy E iso during the prompt phase; and (ii) Yonetoku relation between E i,p and isotropic-equivalent luminosity L iso, both measured during the peak flux interval. [ABSTRACT FROM AUTHOR]

This study used artificial intelligence techniques to identify clinical cancer biomarkers for recurrent gastric cancer survivors. From a hospital-based cancer registry database in Taiwan, the datasets of the incidence of recurrence and clinical risk features were included in 2476 gastric cancer survivors. We benchmarked Random Forest using MLP, C4.5, AdaBoost, and Bagging algorithms on metrics and leveraged the synthetic minority oversampling technique (SMOTE) for imbalanced dataset issues, cost-sensitive learning for risk assessment, and SHapley Additive exPlanations (SHAPs) for feature importance analysis in this study. Our proposed Random Forest outperformed the other models with an accuracy of 87.9%, a recall rate of 90.5%, an accuracy rate of 86%, and an F1 of 88.2% on the recurrent category by a 10-fold cross-validation in a balanced dataset. We identified clinical features of recurrent gastric cancer, which are the top five features, stage, number of regional lymph node involvement, Helicobacter pylori, BMI (body mass index), and gender; these features significantly affect the prediction model's output and are worth paying attention to in the following causal effect analysis. Using an artificial intelligence model, the risk factors for recurrent gastric cancer could be identified and cost-effectively ranked according to their feature importance. In addition, they should be crucial clinical features to provide physicians with the knowledge to screen high-risk patients in gastric cancer survivors as well. [ABSTRACT FROM AUTHOR]

Simple Summary: The differential diagnosis between uterine leiomyosarcomas and leiomyomas based on imaging represents one of the major challenges for gynecologists and radiologists. Currently, only histological examination can definitively resolve doubts in suspicious cases. The purpose of this study is to develop a machine learning model that can support clinical decision making. One of the proposed approaches, i.e., using the general linear model (GLM) classifier, has been patented by our team and has demonstrated good performance in retrospective analyses, with predicted area under the curve (AUC), sensitivity, and specificity on the test set ranging from 0.78 to 0.82, from 0.78 to 0.89, and from 0.67 to 0.87, respectively. The next step will involve validation at other medical centers and its prospective application. Background: The accurate discrimination of uterine leiomyosarcomas and leiomyomas in a pre-operative setting remains a current challenge. To date, the diagnosis is made by a pathologist on the excised tumor. The aim of this study was to develop a machine learning algorithm using radiomic data extracted from contrast-enhanced computed tomography (CECT) images that could accurately distinguish leiomyosarcomas from leiomyomas. Methods: Pre-operative CECT images from patients submitted to surgery with a histological diagnosis of leiomyoma or leiomyosarcoma were used for the region of interest identification and radiomic feature extraction. Feature extraction was conducted using the PyRadiomics library, and three feature selection methods combined with the general linear model (GLM), random forest (RF), and support vector machine (SVM) classifiers were built, trained, and tested for the binary classification task (malignant vs. benign). In parallel, radiologists assessed the diagnosis with or without clinical data. Results: A total of 30 patients with leiomyosarcoma (mean age 59 years) and 35 patients with leiomyoma (mean age 48 years) were included in the study, comprising 30 and 51 lesions, respectively. Out of nine machine learning models, the three feature selection methods combined with the GLM and RF classifiers showed good performances, with predicted area under the curve (AUC), sensitivity, and specificity ranging from 0.78 to 0.97, from 0.78 to 1.00, and from 0.67 to 0.93, respectively, when compared to the results obtained from experienced radiologists when blinded to the clinical profile (AUC = 0.73 95%CI = 0.62–0.84), as well as when the clinical data were consulted (AUC = 0.75 95%CI = 0.65–0.85). Conclusions: CECT images integrated with radiomics have great potential in differentiating uterine leiomyomas from leiomyosarcomas. Such a tool can be used to mitigate the risks of eventual surgical spread in the case of leiomyosarcoma and allow for safer fertility-sparing treatment in patients with benign uterine lesions. [ABSTRACT FROM AUTHOR]

Background: Gene network information is believed to be beneficial for disease module and pathway identification, but has not been explicitly utilized in the standard random forest (RF) algorithm for gene expression data analysis. We investigate the performance of a network-guided RF where the network information is summarized into a sampling probability of predictor variables which is further used in the construction of the RF. Results: Our simulation results suggest that network-guided RF does not provide better disease prediction than the standard RF. In terms of disease gene discovery, if disease genes form module(s), network-guided RF identifies them more accurately. In addition, when disease status is independent from genes in the given network, spurious gene selection results can occur when using network information, especially on hub genes. Our empirical analysis on two balanced microarray and RNA-Seq breast cancer datasets from The Cancer Genome Atlas (TCGA) for classification of progesterone receptor (PR) status also demonstrates that network-guided RF can identify genes from PGR-related pathways, which leads to a better connected module of identified genes. Conclusions: Gene networks can provide additional information to aid the gene expression analysis for disease module and pathway identification. But they need to be used with caution and validation on the results need to be carried out to guard against spurious gene selection. More robust approaches to incorporate such information into RF construction also warrant further study. [ABSTRACT FROM AUTHOR]

Background: Atherosclerosis (AS) is a primary contributor to cardiovascular disease, leading to significant global mortality rates. Developing effective diagnostic indicators and models for AS holds the potential to substantially reduce the fatalities and disabilities associated with cardiovascular disease. Blood sample analysis has emerged as a promising avenue for facilitating diagnosis and assessing disease prognosis. Nonetheless, it lacks an accurate model or tool for AS diagnosis. Hence, the principal objective of this study is to develop a convenient, simple, and accurate model for the early detection of AS. Methods: We downloaded the expression data of blood samples from GEO databases. By dividing the mean values of housekeeping genes (meanHGs) and applying the comBat function, we aimed to reduce the batch effect. After separating the datasets into training, evaluation, and testing sets, we applied differential expression analyses (DEA) between AS and control samples from the training dataset. Then, a gradient-boosting model was used to evaluate the importance of genes and identify the hub genes. Using different machine learning algorithms, we constructed a prediction model with the highest accuracy in the testing dataset. Finally, we make the machine learning models publicly accessible by shiny app construction. Results: Seven datasets (GSE9874, GSE12288, GSE20129, GSE23746, GSE27034, GSE90074, and GSE202625), including 403 samples with AS and 325 healthy subjects, were obtained by comprehensive searching and filtering by specific requirements. The batch effect was successfully removed by dividing the meanHGs and applying the comBat function. 331 genes were found to be related to atherosclerosis by the DEA analysis between AS and health samples. The top 6 genes with the highest importance values from the gradient boosting model were identified. Out of the seven machine learning algorithms tested, the random forest model exhibited the most impressive performance in the testing datasets, achieving an accuracy exceeding 0.8. While the batch effect reduction analysis in our study could have contributed to the increased accuracy values, our comparison results further highlight the superiority of our model over the genes provided in published studies. This underscores the effectiveness of our approach in delivering superior predictive performance. The machine-learning models were then uploaded to the Shiny app's server, making it easy for users to distinguish AS samples from normal samples. Conclusions: A prognostic Shiny application, built upon six potential atherosclerosis-associated genes, has been developed, offering an accurate diagnosis of atherosclerosis. [ABSTRACT FROM AUTHOR]

Software development effort estimation (SDEE) is recognized as vital activity for effective project management since under or over estimating can lead to unsuccessful utilization of project resources. Machine learning (ML) algorithms are largely contributing in SDEE domain, particularly ensemble effort estimation (EEE) works well in rectifying bias and subjectivity to solo ML learners. Performance of EEE significantly depends on hyperparameter composition as well as weight assignment mechanism of solo learners. However, in EEE domain, impact of optimization in terms of hyperparameter tunning as well as weight assignment is explored by few researchers. This study aims in improving SDEE performance by incorporating metaheuristic hyperparameter and weight optimization in EEE, which enables accuracy and diversity to the ensemble model. The study proposed Metaheuristic-optimized Multi-dimensional bagging scheme and Weighted Ensemble (MoMdbWE) approach. This is achieved by proposed search space division and hyperparameter optimization method named as Multi-dimensional bagging (Mdb). Metaheuristic algorithm considered for this work is Firefly algorithm (FFA), to get best hyperparameters of three base ML algorithms (Random Forest, Support vector machine and Deep Neural network) since FFA has shown promising results of fitness in terms of MAE. Further enhancement in performance is achieved by incorporating FFA-based weight optimization to construct Metaheuristic-optimized weighted ensemble (MoWE) of individual multi-dimensional bagging schemes. Proposed scheme is implemented on eight frequently utilized effort estimation datasets and results are evaluated by 5 error metrices (MAE, RMSE, MMRE, MdMRE, Pred), standard accuracy and effect size along with Wilcox statistical test. Findings confirmed that the use of FFA optimization for hyperparameter (with search space sub-division) and for ensemble weights, has significantly enhanced performance in comparison with individual base algorithms as well as other homogeneous and heterogenous EEE techniques. [ABSTRACT FROM AUTHOR]

Sign language recognition presents significant challenges due to the intricate nature of hand gestures and the necessity to capture fine-grained details. In response to these challenges, a novel approach is proposed—Lightweight Attentive VGG16 with Random Forest (LAVRF) model. LAVRF introduces a refined adaptation of the VGG16 model integrated with attention modules, complemented by a Random Forest classifier. By streamlining the VGG16 architecture, the Lightweight Attentive VGG16 effectively manages complexity while incorporating attention mechanisms that dynamically concentrate on pertinent regions within input images, resulting in enhanced representation learning. Leveraging the Random Forest classifier provides notable benefits, including proficient handling of high-dimensional feature representations, reduction of variance and overfitting concerns, and resilience against noisy and incomplete data. Additionally, the model performance is further optimized through hyperparameter optimization, utilizing the Optuna in conjunction with hill climbing, which efficiently explores the hyperparameter space to discover optimal configurations. The proposed LAVRF model demonstrates outstanding accuracy on three datasets, achieving remarkable results of 99.98%, 99.90%, and 100% on the American Sign Language, American Sign Language with Digits, and NUS Hand Posture datasets, respectively. [ABSTRACT FROM AUTHOR]

Predictive modeling often ignores interaction effects among predictors in high-dimensional data because of analytical and computational challenges. Research in interaction selection has been galvanized along with methodological and computational advances. In this study, we aim to investigate the performance of two types of predictive algorithms that can perform interaction selection. Specifically, we compare the predictive performance and interaction selection accuracy of both penalty-based and tree-based predictive algorithms. Penalty-based algorithms included in our comparative study are the regularization path algorithm under the marginality principle (RAMP), the least absolute shrinkage selector operator (LASSO), the smoothed clipped absolute deviance (SCAD), and the minimax concave penalty (MCP). The tree-based algorithms considered are random forest (RF) and iterative random forest (iRF). We evaluate the effectiveness of these algorithms under various regression and classification models with varying structures and dimensions. We assess predictive performance using the mean squared error for regression and accuracy, sensitivity, specificity, balanced accuracy, and F1 score for classification. We use interaction coverage to judge the algorithm's efficacy for interaction selection. Our findings reveal that the effectiveness of the selected algorithms varies depending on the number of predictors (data dimension) and the structure of the data-generating model, i.e., linear or nonlinear, hierarchical or non-hierarchical. There were at least one or more scenarios that favored each of the algorithms included in this study. However, from the general pattern, we are able to recommend one or more specific algorithm(s) for some specific scenarios. Our analysis helps clarify each algorithm's strengths and limitations, offering guidance to researchers and data analysts in choosing an appropriate algorithm for their predictive modeling task based on their data structure. [ABSTRACT FROM AUTHOR]

Predicting crop yield timely can considerably accelerate agricultural production management and food policymaking, which are also important requirements for precise agricultural development. Given the development of hyperspectral imaging technology, a simple and efficient modeling method is convenient for predicting crop yield by using airborne hyperspectral images. In this study, the Unmanned Aerial Vehicle (UAV) hyperspectral and maturity yield data in 2014-2015 and 2017-2018 were collected. The winter wheat yield prediction model was established by optimizing Vegetation Indices (VIs) feature scales and sample scales, incorporating Partial Least Squares Regression (PLSR), Random Forest algorithm (RF), and Back Propagation Neural Network algorithm (BPN). Results showed that PLSR stands out as the optimal wheat yield prediction model considering stability and accuracy (RMSE=948.88 kg/hm²). Contrary to the belief that more input features result in higher accuracy, PLSR, RF, and BPN models performed best when trained with the top 3, 8, and 4 VIs with the highest correlation, respectively. With an increase in training samples, model accuracy improves, reaching stability when the training samples reach 70. Using PLSR and optimal feature scales, UAV yield prediction maps were generated, holding significant value for field management in precision agriculture. [ABSTRACT FROM AUTHOR]

Self-oscillatory systems have great utility in energy harvesting, engines, and actuators due to their ability to convert ambient energy directly into mechanical work. This characteristic makes their design and implementation highly valuable. Due to the complexity of the motion process and the simultaneous influence of multiple parameters, computing self-oscillatory systems proves to be challenging, especially when conducting inverse parameter design. To simplify the computational process, a combined approach o0f Random Forest (RF) and Backpropagation Neural Network (BPNN) algorithms is employed. The example used is a self-rotating skipping rope made of liquid crystal elastomer (LCE) fiber and a mass block under illumination. Numerically solving the governing equations yields precise solutions for the rotation frequency of the LCE skipping rope under various system parameters. A database containing 138,240 sets of parameter conditions and their corresponding rotation frequencies is constructed to train the RF and BPNN models. The training outcomes indicate that RF and BPNN can accurately predict the self-rotating skipping rope frequency under various parameters, demonstrating high stability and computational efficiency. This approach allows us to discover the influences of distinct parameters on the rotation frequency as well. Moreover, it is capable of inverse design, meaning it can derive the corresponding desired parameter combination from a given rotation frequency. Through this study, a deeper understanding of the dynamic behavior of self-oscillatory systems is achieved, offering a new approach and theoretical foundation for their implementation and construction. [ABSTRACT FROM AUTHOR]

In order to study the distribution characteristics of soil organic carbon (SOC) and soil organic carbon storage (SOCS) among different wetland types in Dandou Sea tidal wetland in Guangxi, firstly, based on Sentinel–2 imaging and random forest algorithm, combined with the existing tidal wetland data, a 10 m resolution tidal wetland dataset in Guangxi from 2019 to 2023 was generated, covering mangroves, salt marshes and tidal flats. The results show that the overall accuracy of the recognition results is higher than 96%, and the Kappa coefficient is higher than 0.95, which indicates high accuracy. Subsequently, the distribution characteristics and influencing factors of SOC and SOCS in different habitats were analyzed. The results showed that the SOC content of mangroves and salt marshes was higher than that of tidal flats. The SOC content of mangrove, salt marshes and tidal flats in 0–60 cm soil layer was 5.30–10.42 g/kg, 7.60–9.84 g/kg, and 1.29–2.25 g/kg, respectively. The changes of SOCS were 12.41–26.48 t/ha, 19.58–24.15 t/ha, and 3.61–6.86 t/ha, respectively. With the increase of soil depth, the SOC and SOCS of mangroves decreased gradually, and the SOC and SOCS of salt marshes increased gradually, and SOC and SOCS were mainly affected by soil bulk density (BD), soil moisture content (MC) and pH. [ABSTRACT FROM AUTHOR]

Various ground-based observing techniques provide precipitable water vapor (PWV) products with different spatial resolutions. To effectively integrate these products, especially in terms of vertical orientation, spatial interpolation is essential. In this context, we have developed a model to characterize PWV variation with altitude over our study area. Our model, known as RF-PWV (a PWV vertical correction grid model with a 1° × 1° resolution), is constructed using random forest based on the relationship between the differences in different pressure level PWV data from the fifth-generation European Centre for Medium-Range Weather Forecasts reanalysis (ERA5) monthly average hourly data and corresponding differences in their height differences over time. When validated against 1 h ERA5 PWV profiles, RF-PWV exhibits a 99.84 % reduction in bias and a 63.41 % decrease in the RMSE compared with the most recent model, C-PWVC1. Furthermore, when validated against radiosonde data, RF-PWV shows a 96.36 % reduction in bias and a 5 % decrease in the RMSE compared with C-PWVC1. Additionally, RF-PWV outperforms C-PWVC1 in terms of resistance to seasonal and height difference interference. The model eliminates the need for meteorological parameters, allowing for high-precision PWV vertical correction by inputting only time and height differences. Consequently, RF-PWV can significantly reduce errors in vertical correction, enhance PWV fusion product accuracy, and provide insights into PWV vertical distribution, thereby contributing to climate research. [ABSTRACT FROM AUTHOR]

For shallow lakes, the most dramatic ecosystem shift is that from a clear-water, macrophyte-dominated regime to a turbid, phytoplankton-dominated regime. Whereas many studies have focussed on the factors that trigger such shifts, it is still unclear how these changes shape fish diversity. In the present study, we characterised the fish communities taxonomically and functionally in Lake Taihu, China, along a spatial gradient of habitat types, i.e. algal-dominated, macrophyte-dominated, open-lake and aquaculture conversion zones. Random Forests regression was then applied to examine how fish diversities were related to environmental conditions. Our results showed that the taxonomic and functional composition of the fish community varied among different habitat regimes. Macrophyte and water quality were identified as the most important factors affecting the fish diversity indices. Habitats with a high macrophyte biomass and low nutrient concentration supported higher functional diversity. Moreover, our analysis revealed that although functional diversity indices were strongly correlated with species richness, they appeared more sensitive to changes in habitat conditions. Overall, our study highlights the importance of macrophyte-dominated state in maintaining fish diversity in freshwater ecosystems and provides new insight into how functional diversity can improve our understanding of the response of biodiversity to ecological processes. [ABSTRACT FROM AUTHOR]

Who interacts with whom is a key question in community and network ecology. The concept that these interactions may be driven by a match between the traits of consumer and resource species is known as trait‐matching. If trait‐matching would allow for general predictions of interaction structure based on sufficiently few and easily‐measurable traits, then this approach could replace the laborious description of each individual pairwise interaction. To resolve imprints of trait‐matching in a species‐rich tri‐trophic Salix–galler–parasitoid network, and to identify the most relevant traits, we applied five different methods, each approaching the same phenomenon from a different perspective. As traits, we used, body sizes, gall type (position on plant, structure of gall) and phenology, among others, as well as phylogenetic proxies. When jointly applied, the methods demonstrate distinctly different imprints of traits within the two bipartite network elements (Salix–galler versus galler–parasitoid interactions). Of the galler–parasitoid sub‐network's interactions, approximately half were explainable by the species traits used; of the Salix–galler sub‐network's interactions, traits explained at most two‐fifths. Gall type appeared to be the most important structuring trait in both networks. Phylogeny explained as much, or more than did our tested traits, suggesting that traits may be conserved and phylogeny therefore an effective proxy. Overall, the more specialized structure of the Salix–galler network versus the more nested structure of the galler–parasitoid network meant that different methods were more effective at capturing interactions and interaction structure in the different sub‐networks. Thus, our analysis reveals how structuring impacts may vary even between levels within the same multitrophic network, and calls for comparative analyses of trait matching across a wide set of systems and methods. [ABSTRACT FROM AUTHOR]

The aggregation-induced emission (AIE) effect exhibits a significant influence on the development of luminescent materials and has made remarkable progress over the past decades. The advancement of high-performance AIE materials requires fast and accurate predictions of their photophysical properties, which is impeded by the inherent limitations of quantum chemical calculations. In this work, we present an accurate machine learning approach for the fast predictions of quantum yields and wavelengths to screen out AIE molecules. A database of about 563 organic luminescent molecules with quantum yields and wavelengths in the monomeric/aggregated states was established. Individual/combined molecular fingerprints were selected and compared elaborately to attain appropriate molecular descriptors. Different machine learning algorithms combined with favorable molecular fingerprints were further screened to achieve more accurate prediction models. The simulation results indicate that combined molecular fingerprints yield more accurate predictions in the aggregated states, and random forest and gradient boosting regression algorithms show the best predictions in quantum yields and wavelengths, respectively. Given the successful applications of machine learning in quantum yields and wavelengths, it is reasonable to anticipate that machine learning can serve as a complementary strategy to traditional experimental/theoretical methods in the investigation of aggregation-induced luminescent molecules to facilitate the discovery of luminescent materials. [ABSTRACT FROM AUTHOR]

This study optimized friction stir welding (FSW) parameters for 1.6 mm thick 2024T3 aluminum alloy sheets. A 3 × 3 factorial design was employed to explore tool rotation speeds (1100 to 1300 rpm) and welding speeds (140 to 180 mm/min). Static tensile tests revealed the joints' maximum strength at 87% relative to the base material. Hyperparameter optimization was conducted for machine learning (ML) models, including random forest and XGBoost, and multilayer perceptron artificial neural network (MLP-ANN) models, using grid search. Welding parameter optimization and extrapolation were then carried out, with final strength predictions analyzed using response surface methodology (RSM). The ML models achieved over 98% accuracy in parameter regression, demonstrating significant effectiveness in FSW process enhancement. Experimentally validated, optimized parameters resulted in an FSW joint efficiency of 93% relative to the base material. This outcome highlights the critical role of advanced analytical techniques in improving welding quality and efficiency. [ABSTRACT FROM AUTHOR]

Dynamic wireless charging (DWC) has emerged as a viable approach to mitigate range anxiety by ensuring continuous and uninterrupted charging for electric vehicles in motion. DWC systems rely on the length of the transmitter, which can be categorized into long-track transmitters and segmented coil arrays. The segmented coil array, favored for its heightened efficiency and reduced electromagnetic interference, stands out as the preferred option. However, in such DWC systems, the need arises to detect the vehicle's position, specifically to activate the transmitter coils aligned with the receiver pad and de-energize uncoupled transmitter coils. This paper introduces various machine learning algorithms for precise vehicle position determination, accommodating diverse ground clearances of electric vehicles and various speeds. Through testing eight different machine learning algorithms and comparing the results, the random forest algorithm emerged as superior, displaying the lowest error in predicting the actual position. [ABSTRACT FROM AUTHOR]

This paper focuses on the emissions of the three most sold categories of light vehicles: sedans, SUVs, and pickups. The research is carried out through an innovative methodology based on GPS and machine learning in real driving conditions. For this purpose, driving data from the three best-selling vehicles in Ecuador are acquired using a data logger with GPS included, and emissions are measured using a PEMS in six RDE tests with two standardized routes for each vehicle. The data obtained on Route 1 are used to estimate the gears used during driving using the K-means algorithm and classification trees. Then, the relative importance of driving variables is estimated using random forest techniques, followed by the training of ANNs to estimate CO2, CO, NOX, and HC. The data generated on Route 2 are used to validate the obtained ANNs. These models are fed with a dataset generated from 324, 300, and 316 km of random driving for each type of vehicle. The results of the model were compared with the IVE model and an OBD-based model, showing similar results without the need to mount the PEMS on the vehicles for long test drives. The generated model is robust to different traffic conditions as a result of its training and validation using a large amount of data obtained under completely random driving conditions. [ABSTRACT FROM AUTHOR]

In recent years, soft robotic sensors have rapidly advanced to endow robots with the ability to interact with the external environment. Here, we propose a polymer optical fiber (POF) sensor with sensitive and stable detection performance for strain, bending, twisting, and pressing. Thus, we can map the real-time output light intensity of POF sensors to the spatial morphology of the elastomer. By leveraging the intrinsic correlations of neighboring sensors and machine learning algorithms, we realize the spatially resolved detection of the pressing and multi-dimensional deformation of elastomers. Specifically, the developed intelligent sensing system can effectively recognize the two-dimensional indentation position with a prediction accuracy as large as ~99.17%. The average prediction accuracy of combined strain and twist is ~98.4% using the random forest algorithm. In addition, we demonstrate an integrated intelligent glove for the recognition of hand gestures with a high recognition accuracy of 99.38%. Our work holds promise for applications in soft robots for interactive tasks in complex environments, providing robots with multidimensional proprioceptive perception. And it also can be applied in smart wearable sensing, human prosthetics, and human–machine interaction interfaces. [ABSTRACT FROM AUTHOR]

Wearable sensors could be beneficial for the continuous quantification of upper limb motor symptoms in people with Parkinson's disease (PD). This work evaluates the use of two inertial measurement units combined with supervised machine learning models to classify and predict a subset of MDS-UPDRS III subitems in PD. We attached the two compact wearable sensors on the dorsal part of each hand of 33 people with PD and 12 controls. Each participant performed six clinical movement tasks in parallel with an assessment of the MDS-UPDRS III. Random forest (RF) models were trained on the sensor data and motor scores. An overall accuracy of 94% was achieved in classifying the movement tasks. When employed for classifying the motor scores, the averaged area under the receiver operating characteristic values ranged from 68% to 92%. Motor scores were additionally predicted using an RF regression model. In a comparative analysis, trained support vector machine models outperformed the RF models for specific tasks. Furthermore, our results surpass the literature in certain cases. The methods developed in this work serve as a base for future studies, where home-based assessments of pharmacological effects on motor function could complement regular clinical assessments. [ABSTRACT FROM AUTHOR]

The availability of a sufficient number of annotated samples is one of the main challenges of the supervised methods used to classify crop types from remote sensing images. Creating these samples is time-consuming and costly. Active Learning (AL) offers a solution by streamlining sample annotation, resulting in more efficient training with less effort. Unfortunately, most of the developed AL methods overlook spatial information inherent in remote sensing images. We propose a novel spatially explicit AL that uses the semi-variogram to identify and discard redundant, spatially adjacent samples. It was evaluated using Random Forest (RF) and Sentinel-2 Satellite Image Time Series in two study areas from the Netherlands and Belgium. In the Netherlands, the spatially explicit AL selected 97 samples achieving an overall accuracy of 80%, compared to traditional AL selecting 169 samples with 82% overall accuracy. In Belgium, spatially explicit AL selected 223 samples and obtained 60% overall accuracy, while traditional AL selected 327 samples and obtained an overall accuracy of 63%. We concluded that the developed AL method helped RF achieve a good performance mostly for the classes consisting of individual crops with a relatively distinctive growth pattern such as sugar beets or cereals. Aggregated classes such as 'fruits and nuts' posed, however, a challenge. [ABSTRACT FROM AUTHOR]

Objective: This paper proposes an objective method to measure and identify trust-change directions during takeover transitions (TTs) in conditionally automated vehicles (AVs). Background: Takeover requests (TORs) will be recurring events in conditionally automated driving that could undermine trust, and then lead to inappropriate reliance on conditionally AVs, such as misuse and disuse. Method: 34 drivers engaged in the non-driving-related task were involved in a sequence of takeover events in a driving simulator. The relationships and effects between drivers' physiological responses, takeover-related factors, and trust-change directions during TTs were explored by the combination of an unsupervised learning algorithm and statistical analyses. Furthermore, different typical machine learning methods were applied to establish recognition models of trust-change directions during TTs based on takeover-related factors and physiological parameters. Result: Combining the change values in the subjective trust rating and monitoring behavior before and after takeover can reliably measure trust-change directions during TTs. The statistical analysis results showed that physiological parameters (i.e., skin conductance and heart rate) during TTs are negatively linked with the trust-change directions. And drivers were more likely to increase trust during TTs when they were in longer TOR lead time, with more takeover frequencies, and dealing with the stationary vehicle scenario. More importantly, the F1-score of the random forest (RF) model is nearly 77.3%. Conclusion: The features investigated and the RF model developed can identify trust-change directions during TTs accurately. Application: Those findings can provide additional support for developing trust monitoring systems to mitigate both drivers' overtrust and undertrust in conditionally AVs. [ABSTRACT FROM AUTHOR]

Background: Shortages of mechanical ventilation have become a constant problem in Emergency Departments (EDs), thereby affecting the timely deployment of medical interventions that counteract the severe health complications experienced during respiratory disease seasons. It is then necessary to count on agile and robust methodological approaches predicting the expected demand loads to EDs while supporting the timely allocation of ventilators. In this paper, we propose an integration of Artificial Intelligence (AI) and Discrete-event Simulation (DES) to design effective interventions ensuring the high availability of ventilators for patients needing these devices. Methods: First, we applied Random Forest (RF) to estimate the mechanical ventilation probability of respiratory-affected patients entering the emergency wards. Second, we introduced the RF predictions into a DES model to diagnose the response of EDs in terms of mechanical ventilator availability. Lately, we pretested two different interventions suggested by decision-makers to address the scarcity of this resource. A case study in a European hospital group was used to validate the proposed methodology. Results: The number of patients in the training cohort was 734, while the test group comprised 315. The sensitivity of the AI model was 93.08% (95% confidence interval, [88.46 − 96.26%]), whilst the specificity was 85.45% [77.45 − 91.45%]. On the other hand, the positive and negative predictive values were 91.62% (86.75 − 95.13%) and 87.85% (80.12 − 93.36%). Also, the Receiver Operator Characteristic (ROC) curve plot was 95.00% (89.25 − 100%). Finally, the median waiting time for mechanical ventilation was decreased by 17.48% after implementing a new resource capacity strategy. Conclusions: Combining AI and DES helps healthcare decision-makers to elucidate interventions shortening the waiting times for mechanical ventilators in EDs during respiratory disease epidemics and pandemics. [ABSTRACT FROM AUTHOR]

Tropical forests are the world's most structurally complex ecosystems, providing key functions like biomass accumulation, which is linked to this complexity. Tropical forests are also exposed to chronic, non-severe winds, yet their effect on forest structural complexity is understudied. Here we examine drivers of forest structural complexity in Puerto Rico with a particular focus on chronic wind exposure. We used airborne light detection and ranging data collected in 2016 to quantify canopy height and rugosity (variation in height) in ~ 20,000, 0.28 ha forested sites stratified by forest age. We used random forest models to analyze variation in canopy height and rugosity as a function of chronic wind exposure, forest age, mean annual precipitation, elevation, slope (in degrees), soil type, soil available water storage, and exposure to a previous hurricane. Canopy height was driven by precipitation, forest age, and chronic wind exposure, decreasing by 2.12 m (16%) on average in wind-exposed forests across all forest ages. Canopy height increased by 4.0 m (41%) on average in forests aged 25–66 years, and by 4.0 m between sites with 1000 and 2000 mm y−1 precipitation. Canopy rugosity was driven by canopy height, precipitation, forest age, and elevation, increasing log-linearly with canopy height and precipitation, decreasing with elevation, and was highest in younger forests. Chronic wind exposure did not drive variation in canopy rugosity. Our results suggest that chronic wind exposure plays an integral role in limiting canopy height, potentially reducing aboveground carbon accumulation in older tropical forests. [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]

The article proposes to use machine learning as one of the areas of artificial intelligence to forecast the volume of biogas production from household organic waste. The use of five regression algorithms (Linear Regression, Ridge Regression, Lasso Regression, Random Forest Regression, and Gradient Boosting Regression) to create an effective model for forecasting the volume of biogas production from household organic waste is considered. Based on the comparison of these algorithms by MSE and MAE indicators, the quality of training and their accuracy during forecasting are evaluated. The proposed algorithm for creating a model for forecasting biogas production volumes from household organic waste involves the implementation of 10 main and 3 auxiliary steps. Their advantage is that they aid in the performance of component data analysis, which is carried out based on the method of reducing the dimensionality of the data set, increasing interpretability, and minimizing the risk of data loss. An analysis of 2433 data is was carried out, which characterizes the formation of biogas from food (FW) and yard waste (YW) according to four features. Data preparation is performed using the Jupyter Notebook environment in Python. We select five machine learning algorithms to substantiate an effective model for forecasting volumes of biogas production from household organic waste. On the basis of the conducted research, the main advantages and disadvantages of the used algorithms for building forecasting models of biogas production volumes from household organic waste are determined. It is found that two models, "Random Forest Regressor" and "Gradient Boosting Regressor", show the best accuracy indicators. The other three models (Linear Regression, Ridge Regression, Lasso Regression) are inferior in accuracy and were not considered further. To determine the accuracy of the "Random Forest Regressor" and "Gradient Boosting Regressor" models, we choose the MSE and MAE indicators. The Random Forest Regressor model is found to be a more accurate model compared to the Gradient Boosting Regressor. This is confirmed by the fact that the MSE of the "Random Forest Regressor" model on the training data set is 7.14 times smaller than that of the "Gradient Boosting Regressor" model. At the same time, MAE is 2.67 times smaller in the "Random Forest Regressor" model than in the "Gradient Boosting Regressor" model. The MSE and MAE of both models are worse on the test data set, which indicates overtraining tendencies. The Gradient Boosting Regressor model has worse MSE and MAE than the Random Forest Regressor model on both the training and test data sets. It is established that the model based on the "Random Forest Regressor" algorithm is the most effective for forecasting the volume of biogas production from household organic waste. It provides MAE = 0.088 on test data and the smallest absolute errors in predictions. Further systematic improvement of the "Random Forest Regressor" model for forecasting biogas production volumes from household organic waste based on new data will ensure its accuracy and maintain competitive advantages. [ABSTRACT FROM AUTHOR]

Southern China, one of the traditional rice production bases, has experienced significant declines in the area of rice paddy since the beginning of this century. Monitoring the rice cropping area is becoming an urgent need for food security policy decisions. One of the main challenges for mapping rice in this area is the quantity of cloud-free observations that are vulnerable to frequent cloud cover. Another relevant issue that needs to be addressed is determining how to select the appropriate classifier for mapping paddy rice based on the cloud-masked observations. Therefore, this study was organized to quickly find a strategy for rice mapping by evaluating cloud-mask algorithms and machine-learning methods for Sentinel-2 imagery. Specifically, we compared four GEE-embedded cloud-mask algorithms (QA60, S2cloudless, CloudScore, and CDI (Cloud Displacement Index)) and analyzed the appropriateness of widely accepted machine-learning classifiers (random forest, support vector machine, classification and regression tree, gradient tree boost) for cloud-masked imagery. The S2cloudless algorithm had a clear edge over the other three algorithms based on its overall accuracy in evaluation and visual inspection. The findings showed that the algorithm with a combination of S2cloudless and random forest showed the best performance when comparing mapping results with field survey data, referenced rice maps, and statistical yearbooks. In general, the research highlighted the potential of using Sentinel-2 imagery to map paddy rice with multiple combinations of cloud-mask algorithms and machine-learning methods in a cloud-prone area, which has the potential to broaden our rice mapping strategies. [ABSTRACT FROM AUTHOR]

Precipitation is the basic component of the Earth's water cycle. Obtaining high-resolution and high-precision precipitation data is of great significance. This paper establishes a precipitation retrieval model based on a random forest classification and regression model during the day and at night with FY-4B/AGRI Level1 data on China from July to August 2022. To evaluate the retrieval effect of the model, the GPM IMERG product is used as a reference, and the retrieval results are compared against those of the FY-4B/AGRI operational precipitation product. In addition, the retrieval results are analyzed according to different underlying surfaces. The results showed that compared with the FY-4B/AGRI operational precipitation product, the retrieval model can better identify precipitation and capture precipitation areas of light rain, moderate rain, heavy rain and torrential rain. Among them, the probability of detection (POD) of the day model increased from 0.328 to 0.680, and the equitable threat score (ETS) increased from 0.252 to 0.432. The POD of the night model increased from 0.337 to 0.639, and the ETS score increased from 0.239 to 0.369. Meanwhile, the precipitation estimation accuracy of the day model increased by 38.98% and that of the night model increased by 40.85%. Our results also showed that due to the surface uniformity of the ocean, the model can identify precipitation better on the ocean than on the land. Our findings also indicated that for the different underlying surfaces of the land, there is no significant difference in each evaluation index of the model. This is a strong argument for the universal applicability of the model. Notably, the results showed that, especially for more vegetated areas and areas covered by water, the model is capable of estimating precipitation. In conclusion, the precipitation retrieval model that is proposed herein can better determine precipitation regions and estimate precipitation intensities compared with the FY-4B/AGRI operational precipitation product. It can provide some reference value for future precipitation retrieval research on FY-4B/AGRI. [ABSTRACT FROM AUTHOR]