Search

Identifying the most reliable reanalysis temperature products is crucial for advancing hydro-climate research in data-scarce regions. This study evaluated two widely used reanalysis datasets in estimating minimum temperature (Tmin) and maximum temperature (Tmax) across various Agro-Ecological Zones (AEZs) of Ethiopia at different temporal scales over the period 1990 to 2020. These datasets include the Modern-Era Retrospective Analysis for Research and Applications, version 2.0 (MERRA v2.0) and the fifth generation of the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis (ERA5). Evaluations was undertaken at each AEZs using five statistical metrics, scatter plots and line graphs by the data collected from 37 meteorological stations. The results indicate that MERRA v2.0 demonstrates superior performance in simulating Tmin across daily, monthly, and annual temporal scales in most AEZs, based on the majority or entirety of the evaluation metrics. For Tmax, MERRA v2.0 performs best on an annual scale, while ERA5 shows better performance on a daily scale. However, both reanalysis products exhibit comparable performance in estimating monthly Tmax. Notably, the performance of the reanalysis datasets varies across different AEZs. In general, the findings underscore the importance of selecting the most suitable reanalysis datasets for Tmin and Tmax separately, tailored to specific temporal scales and AEZs, to improve hydrology and/or climate studies in data scarce portions of the country. [ABSTRACT FROM AUTHOR]

By implementing a multi‐objective optimization approach in forecasting, we introduce three optimization models grey wolf optimizer, genetic algorithm, and differential evolution algorithm combined with multilayer perceptron neural networks and support vector machines to predict electricity consumption in the UAE. The hybrid models' accuracy and efficiency were evaluated using various forecasting metrics. This study's contributions are threefold: it is the first to employ such a sophisticated hybrid approach, particularly using the recently introduced grey wolf optimizer, it compares optimization techniques with the established Pearson correlation‐based method for dimensionality reduction and it represents one of the most extensive macroeconomic forecasts in the UAE using multi‐objective heuristic hybrid optimization methods. Our findings indicate that the grey wolf optimizer significantly outperforms all other models, followed by the genetic algorithm. [ABSTRACT FROM AUTHOR]

Ensuring the integrity and confidentiality of patient information is crucial since it contains sensitive information. Normally the source or owners’ information is used as watermark which is susceptible to illegal manipulations. This research work proposes an imperceptible and reversible watermarking algorithm designed to embed image content-based information for detecting illegal alterations in medical images. The algorithm operates by first normalizing the medical image within the range 0 to 255 and dividing it logically into 4×4 equal blocks. Subsequently, for each block, a Gray-level Co-occurrence Matrix is computed, enabling the derivation of statistical correlations such as contrast, correlation, energy, and homogeneity. These statistical measures collectively form a unique signature, which serves as the watermark to embed in the least significant bits of the pixels of the cover image. Further, in order to verify its integrity, a watermark can be extracted and compared with the original. The proposed watermarking method accelerates the accurate recovery of medical images by retaining their originality and also facilitates the efficient detection of unauthorized manipulations. Experimental results demonstrate the superior imperceptibility achieved by the proposed approach. The proposed method can be used with medical imaging modalities such as X-rays, CT scans, MRI scans, ultrasound scans, etc. [ABSTRACT FROM AUTHOR]

This paper introduces a new dissimilarity measure between two discrete and finite probability distributions. The followed approach is grounded jointly on mixtures of probability distributions and an optimization procedure. We discuss the clear interpretation of the constitutive elements of the measure under an information-theoretical perspective by also highlighting its connections with the Rényi divergence of infinite order. Moreover, we show how the measure describes the inefficiency in assuming that a given probability distribution coincides with a benchmark one by giving formal writing of the random interference between the considered probability distributions. We explore the properties of the considered tool, which are in line with those defining the concept of quasimetric—i.e. a divergence for which the triangular inequality is satisfied. As a possible usage of the introduced device, an application to rare events is illustrated. This application shows that our measure may be suitable in cases where the accuracy of the small probabilities is a relevant matter. [ABSTRACT FROM AUTHOR]

Copyright of Przegląd Elektrotechniczny is the property of Przeglad Elektrotechniczny and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The global climate models (GCMs) are sophisticated tools for determining how the climate system will respond. However, the output of GCMs has a coarse resolution, which is unsuitable for basin-level modelling. Global climate models need to be downscaled at a local/basin scale to determine the impacts of climate change on hydrological responses. The present study attempted to evaluate how effectively various large-scale predictors could reproduce local-scale rain in 35 different locations in India using artificial neural networks (ANN), change-factors (CF), K-nearest neighbour (KNN), and multiple linear regression (MLR). The selection of predictors is made based on the correlation value. As potential predictors, air temperature, geo-potential height, wind velocity component, and relative humidity at specific mean sea-level pressure are selected. The comparison of four different downscaling methods concerning the reproduction of various statistics such as mean, standard deviation at chosen locations, quantile–quantile plots, cumulative distribution function, and kernel density estimation of the PDFs of daily rainfall for selected stations is examined. The CF approach outperforms the other methods at almost all sites (R2 = 0.92–0.99, RMSE = 1.37–28.88 mm, and NSE = –16.55–0.99). This also closely resembles the probability distribution pattern of IMD data. [ABSTRACT FROM AUTHOR]

Climate variability is a highly debated and unavoidable global environmental challenge that has adverse effects on Ethiopia, a developing country. Hence, the objective of this research is to examine the changes in rainfall patterns in Addis Ababa City, Ethiopia, from 1981 to 2018, considering both spatial and temporal aspects. The study utilized a time-series dataset of climate information, which had a spatial resolution of 4 × 4 km, obtained from the National Meteorological Agency of Ethiopia. Supplementary data was also acquired from the Ethiopian Space Science and Geospatial Institute. To examine the rainfall variability, statistical measures such as the coefficient of variation (CV) and standardized anomaly index (SAI) were employed. Geospatial technologies and "R" programming were also used to perform a non-parametric Mann-Kendall (MK) test and Sen's slope estimator for the investigation of both the trend and magnitude of changes. The annual, Kiremt (main rainy), and Belg (spring) seasons rainfall exhibited low to moderate variability with CV < 20% and CV < 30%, respectively, and very high variability for the Belg season (CV > 30%). The Bega season's variability was extreme (CV > 70%). In contrast, decadal rainfall variability was generally very low (CV < 10%). The months from October to March showed higher inter-monthly variability, with CV exceeding 100%. In contrast, the Kiremt season, July, and August, experienced lower inter-monthly variability (CV < 30%). The western, north-east, and southern parts of Addis Ababa demonstrated relatively higher rainfall variability, and the trends decreased in all seasons and months, except the Kiremt season and the months of May, June, and September. However, none of these seasonal and monthly changes were statistically significant (P > 0.05). The study identified 6 years (1982, 1984, 1997, 1999, 2014, and 2015) with varying degrees of drought. Consequently, the spatio-temporal variability of precipitation should be considered in development plans, disaster risk reduction strategies, and policy measures such as flood management. [ABSTRACT FROM AUTHOR]

Abstract: Feedback at work refers to the amount and availability of positive and negative feedback at work. The objective was to propose a short version and seek evidence of validity for the Perceived Feedback at Work Scale (FES) in the Brazilian context. The sample consisted of 552 Brazilian workers of both sexes. The results obtained indicated that the 14-item structure, divided equally into two correlated factors (supervisor and coworkers), showed adequate fit indices and was invariant between the groups divided in terms of modality and organizational sector. Supervisor feedback showed a positive correlation with support from the supervisor, and feedback from colleagues with social support at work. There was also evidence of discriminant validity between the constructs. It is concluded that the proposed version presents adequate psychometric evidence.

This study uses a low-cost Lantana trifolia-activated carbon and an artificial neural network (ANN) to predict wastewater having 2-phenoxyethanoic acid adsorption. To forecast 2-phenoxyethanoic acid elimination efficiency, MATLAB uses a three-layer feed-forward neural network, a Multilayer Perceptron (MLP), with backpropagation. The neural network is trained with three input parameters: 2-phenoxyethanoic acid concentration, adsorbent amount, and contact duration. Root mean square error and linear regression estimate the removal effectiveness of 2-phenoxyethanoic acid. The findings reveal that the generated ANN models for 2-phenoxyethanoic acid adsorption accurately match empirical data on pollutant removal efficacy. This suggests that low-cost Lantana trifolia activated carbon may be used to estimate 2-phenoxyethanoic acid absorption using the neural network technique. [ABSTRACT FROM AUTHOR]

Copyright of Paideia (0103863X) is the property of Paideia and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

As crop yield is determined by numerous input parameters, it is important to identify the most important variables/parameters and eliminate those that may reduce the accuracy of the prediction models. The feature selection algorithms assist in selecting only those relevant features for the prediction algorithms. Instead of a complete set of features, feature subsets give better results for the same algorithm with less computational time. Feature selection has the potential to play an important role in the agriculture domain, with the crop yield depending on multiple factors such as land use, water management, fertilizer application, other management practices and weather parameters. In the present study, feature selection algorithms such as forward selection, backward selection, random forest (RF) and least absolute shrinkage and selection operator (LASSO) have been applied to three different datasets. Regression forecasting models have been developed with selected features for all the algorithms. The forecasting performance of the proposed models was compared using statistical measures such as root mean square error, mean absolute prediction error and mean absolute deviation. A comparison was made among all the feature selection algorithms. The regression models developed with LASSO, RF and backward selection algorithms were the best for different datasets. [ABSTRACT FROM AUTHOR]

The strength of carbon nanotubes (CNTs) and cement composites is dependent on multiple variables. In addition, CNTs added to a cement-based matrix can boost its strength. However, the information related to CNTs characteristics is limited and scarce. Their incorporation may substantially enhance the mechanical and durability properties of cementitious mixtures. Despite challenges such as high cost and workability problems. Therefore, proper consumption of these materials must be used to attain desired qualities. The principal plan of this investigation is to create predictive framework by utilizing machine-learning algorithms. Gene expression programming (GEP), and the random forest algorithm (RFA) is employed to estimate the compressive strength of concrete mixed with CNTs. GEP is used as an individual approach, and RFA is used as an ensemble method to depict the most influential model. The outcomes of the two models are assessed by employing external K-fold cross-validation, and a comparison is done. A comprehensive database is established comprising 282 data points for the CS with blended CNTs. The model is then calibrated using six inputs, including curing time (CT), water-to-cement ratio (W/C), fine aggregate (FA), carbon nanotube content (CNTs), cement content (CC), and coarse aggregate (CA). In addition, the predicted results are validated using k-fold cross-validation, and performance metrics, such as mean absolute error (MAE), root squared error (RSE), correlation coefficient (R2), root mean square error (RMSE), and relative root mean square error (RRMSE). The result shows that RF regression with the nth estimator shows robust accuracy by showing minimal errors as analyzed to individual RF and GEP models. Likewise, the nth model depicts higher R2 = 0.96, and validation results demonstrate low errors. Moreover, the GEP model excels in terms of prediction through the empirical equation. In addition, Shapley analysis (SHAP) is performed to check the distribution of parameters to output. The result reveals that curing time, cement, and water to binder have substantial influence of CNT based concrete composite.

While some robust artificial intelligence (AI) techniques such as Gene-Expression Programming (GEP), Model Tree (MT), and Multivariate Adaptive Regression Spline (MARS) have been frequently employed in the field of water resources, documents aimed to explore their uncertainty levels are few and far between. Meanwhile, uncertainty determination of these AI models in practical applications is highly important especially when we aimed to use the AI models for streamflow forecast due to the repercussions of poorly managed water resources. With the aid of a global daily streamflow dataset, understanding the uncertainty of GEP, MT, and MARS for forecasting streamflow of natural rivers was studied. The efficiency of uncertainty analysis was quantified by two statistical indicators: 95% Percent Prediction Uncertainty (95%PPU) and R-factor. The results demonstrated that MT had lower uncertainty (95%PPU=0.59 and R-factor=1.67) in comparison with MARS (95%PPU=0.61 and R-factor=1.92) and GEP (95%PPU=0.64 and R-factor=2.03). Overall, although the confidence interval bands of uncertainty for the AI models almost captured the mean streamflow measurements, wide bands of uncertainty were obtained and consequently remarkable uncertainty in the calculation of monthly streamflow values was met. [ABSTRACT FROM AUTHOR]

Systemic blood pressure, recorded by various parameters, always shows variation. This is known as 'variability', an entity not assessed routinely. Research has shown greater prognostic importance of it than routine parameters. Various intrinsic and extrinsic factors modulate it. Depending on the interval of successive readings, blood pressure variability is of five types, with different non-invasive methods utilized for recording. It is calculated by various statistical parameters, the most common being standard deviation, but average real variability is the most accurate and easily applied. Clinical evidence is increasing rapidly, indicating variability as a prognostic marker for stroke, ischemic heart disease, renal failure, cognitive dysfunction, heart failure, and mortality. Therapeutic measures for the control of variability have also been forwarded. The paucity of clinical application of blood pressure variability is the stimulus to narrate this review, especially for physicians managing hypertensive patients . [ABSTRACT FROM AUTHOR]

In the field of bioinformatics, the classification of tumors is a difficult and time-consuming task. When diagnosing cancer, gene expression levels are typically one of the most useful tools. However, the biological noise present in microarray data leads to unsatisfactory precision and accuracy. The utilization of thousands of genes in the process of diagnosing tumors is an important task. The two levels of feature selection have been proposed in order to determine the genes that are the most informative to diagnose cancer. Using three different statistical methods, the first level of selection reveals the prognostic genes. In the second level, the differential evolution algorithm considers the prognostic genes that were obtained from statistical measures as initial members to identify the most relevant features. The scaling factor in the modified differential evolution algorithm was made to vary in a dynamic manner in order to evolve the mutant member of the population. The proposed model is a hybrid of statistical approach and evolutionary computation with modified differential evolution algorithm that identifies the candidate genes from thousands of genes from gene expression data. The findings obtained through this hybrid approach upon testing five gene expression datasets provide evidence that it has outperformed when compared to the existing systems for DLBCL outcome, prostate outcome, prostate, and colon tumor datasets with improved classification accuracies of 14%, 4%, 0.62%, and 0.13%, respectively. [ABSTRACT FROM AUTHOR]

Portfolio optimization has always been a challenging proposition in finance and management. Portfolio optimization facilitates in selection of portfolios in a volatile market situation. In this paper, different classical, statistical and intelligent approaches employed for portfolio optimization and management are reviewed. A brief study is performed to understand why portfolio is important for any organization and how recent advances in machine learning and artificial intelligence can help portfolio managers to take right decisions regarding allotment of portfolios. A comparative study of different techniques, first of its kind, is presented in this paper. An effort is also made to compile classical, intelligent, and quantum-inspired techniques that can be employed in portfolio optimization. [ABSTRACT FROM AUTHOR]

Simple Summary: Molecular testing can stratify the risk of malignancy among indeterminate thyroid nodules and subsequently reduce the need for diagnostic surgery. We assessed the performance of the GSC + XA; ThyGeNEXT + ThyraMIR, and GSC + GEC molecular platforms to elucidate their diagnostic accuracy. All three platforms were reliable for ruling out thyroid cancer, with a tendency for the newest Afirma genetic testing platform (GSC + XA) to outperform the other platforms. Indeterminate thyroid nodules (ITN) represent 20–30% of biopsied nodules, with a 10–60% risk of malignancy. Molecular testing can stratify the risk of malignancy among ITNs, and subsequently reduce the need for unnecessary diagnostic surgery. We aimed to assess the performance of these molecular tests at a single institution. Patients with Bethesda III, IV, and V nodules with Afirma and Interpace Diagnostics genetic testing data from November 2013 to November 2021 were included. Three cohorts were formed, including GSC + XA, ThyGeNEXT + ThyraMIR, and GSC + GEC. Statistical analysis determined the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), diagnostic odds ratio (DOR), and accuracy of each type of testing. The PPV of nodules undergoing genetic testing by ThyGeNEXT + ThyraMIR (45.00%, 95%CI: 28.28–62.93%, p = 0.032) and GSC + XA (57.14%, 95%CI: 29.32–81.08%, p < 0.001) were superior to that of GEC + GSC (30.72%, 95%CI: 26.83–34.90%). The NPV was above 85% in all cohorts, suggesting overall suitable rule-out tests. The Afirma platform (GSC + XA) had the highest NPV at 96.97%. The overall accuracy for nodules undergoing ThyGeNEXT + ThyraMIR was 81.42% (95%CI: 73.01–88.11%, p < 0.001). A total of 230 patients underwent thyroidectomy, including less than 60% of each of the ThyGeNEXT + ThyraMIR and GSC + XA cohorts. Specifically, only 25% of patients in the GSC + XA cohort underwent surgery, considerably decreasing the rate of unnecessary surgical intervention. Sub-group analysis, including only patients with surgical pathology, found that PPV tended to be higher in the GSC + XA cohort, at 66.67% (95%CI: 37.28–87.06%), as compared to the ThyGeNEXT + ThyraMIR cohort, at 52.94% (95%CI: 35.25–69.92%). The Afirma genetic testing platform GSC + XA outperformed the other platforms with regards to both PPV and NPV and decreased the rate of surgery in patients with ITNs by 75%, significantly preventing unnecessary surgical intervention. [ABSTRACT FROM AUTHOR]

In this paper, a new generalization of one parameter Lindely distribution is proposed. The new distribution is a mixture distribution of Gamma distributions with fixed scale parameter and variable shape parameter. The distribution is called 'GOLD Distribution' as it is a generalization for several distributions such as exponential, Lindely, Sujatha, Amarendra, Devya and Shambhu distributions. The probability density and cumulative density functions are derived. Also, the statistical properties of the GOLD distribution are discussed. Parameter estimation using the maximum likelihood and the method of moments are given. Moreover, an illustration of the usefulness of the GOLD distribution in survival data analysis is discussed based on a real lifetime data. [ABSTRACT FROM AUTHOR]

This research aims to predict the downward displacement of piles as the axial loading increases and decreases to form the pile static load curves. The generalized regression neural network (GRNN) is used to predict the behavior of continuous flight auger piles through training and testing the network by using field data collected from over 100 field static load tests. The model aims to reduce cost, decrease the dependency on the lengthy field test, and complement the current methods used in pile design. In this research, the authors built a neural network that associates the ground properties, such as friction angle and unconfined compressive strength, as well as the pile's length and diameter with the CFA pile performance at different sites. More than six thousand field points were collected to train the neural network. Moreover, three independent field tests were used to examine the performance of the network comparing the predicted values with the field data. The statistical measures showed that the proposed model can be generalized. The model can predict the pile displacement curves accurately as long as the testing data fall within the training set. If the testing data fall outside the range, the model was not as accurate. To evaluate the performance of the GRNN, three statistical measures were calculated; the coefficient of efficiency ranged between 0.8362 and 0.9968, the root mean square error ranged between 0.0868 and 0.2760, and the mean absolute error ranged between 0.070469 and 0.23443. [ABSTRACT FROM AUTHOR]

Landslide susceptibility prediction can be considered a crucial step in landslide risk assessment. This prediction helps in planning the land use properly. The primary aim of the study is to investigate different machine learning methods and develop anatomy to train and validate the landslide susceptibility prediction models with the help of various statistical techniques. The Kullu–Rohtang pass transport corridor has been selected as the study area. Initially, a landslide inventory was prepared using different sources and nine landslide triggering features were used for further study. All landslide locations in the study area were arbitrarily divided into a ratio of 67:33 to train and test various landslide susceptibility prediction models. The best-triggering features were chosen with the help of the information gain ratio (IGR) defining the predictive capability of different triggering features. Afterwards, five landslide susceptibility prediction models were constructed using a decision tree, K-nearest neighbour (KNN), Gaussian Naïve Bayes, support vector machine (SVM) and multilayer perceptron (MLP). The comparison and validation study of different resulting models was done by applying the receiver operating characteristic (ROC) curve, the kappa index and other statistical methods. Results show that the different models have the outstanding predictive capability with the decision tree model (100%), the Gaussian Naïve Bayes model (100%), the SVM model (100%), and the MLP model (100%) and the KNN model (99.9%). The result indicates statistical differences among various models. The validation results demonstrate the perfect agreement between the expected and predicted landslides along the transport corridor. [ABSTRACT FROM AUTHOR]

A trustworthy assessment of soil moisture content plays a significant role in irrigation planning and in controlling various natural disasters such as floods, landslides, and droughts. Various machine learning models (MLMs) have been used to increase the accuracy of soil moisture content prediction. The present investigation aims to apply MLMs with novel structures for the estimation of daily volumetric soil water content, based on the stacking of the multilayer perceptron (MLP), random forest (RF), and support vector regression (SVR). Two groups of input variables were considered: the first (Model A) consisted of various meteorological variables (i.e., daily precipitation, air temperature, humidity, and wind speed), and the second (Model B) included only daily precipitation. The stacked model (SM) had the best performance (R2 = 0.962) in the prediction of daily volumetric soil water content for both categories of input variables when compared with the MLP (R2 = 0.957), RF (R2 = 0.956) and SVR (R2 = 0.951) models. Overall, the SM, which, in general, allows the weaknesses of the individual basic algorithms to be overcome while still maintaining a limited number of parameters and short calculation times, can lead to more accurate predictions of soil water content than those provided by more commonly employed MLMs. [ABSTRACT FROM AUTHOR]

Recently, many studies have examined filters for reducing or removing speckle noise, which is inherent to different images types such as Porous Silicon (PS) images, in order to ameliorate the metrological evaluation of their applications. In the case of digital images, noise can produce difficulties in the diagnosis of images details, such as edges and limits, should be preserved. Most algorithms can reduce or remove speckle noise, but they do not consider the conservation of these details. This paper describes in detail, the different techniques that focus mainly on the smoothing or elimination of speckle noise in images, as the aim of this study is to achieve the improvement of this smoothing and elimination, which is directly related to different processes (such as the detection of interest regions). Furthermore, the description of these techniques facilitates the operations of evaluations and research with a more specific scope. This study initially covers the definition and modeling of speckle noise. Then we elaborated in detail the different types of filters used in this study, finally, five statistical parameters such as Root Mean Square Error (RMSE), Mean Square Error (MSE), Structural Similarity Index (SSIM), Peak Signal to Noise Ratio (PSNR), Signal to Noise Ratio (SNR) are calculated, compared and the results are tabulated, common in filter evaluation processes. Trough the calculation of the statistical parameters, we can classify the filters in terms of perceptual quality by providing greater certainty. [ABSTRACT FROM AUTHOR]

Traditional approach based on wind tunnel tests used in bridge wind-resistant design is no more efficient since long-span bridges require an iterative process involving multiple design variables and processes. To ease the design process, scholars have proposed aerodynamic shape optimization based on the Kriging surrogate model. Despite the prowess of the Kriging model, it is accurate only when using high-fidelity data (experimental or Large Eddy Simulation data) and is not suitable for noisy datasets, making this model costly and hindering its practical use. To tackle this issue, the present study proposed a polynomial surrogate combined with a genetic algorithm to determine the optimal shape of a streamlined bridge deck cross-section. First, a uniform sampling plan considering 57 different geometries was generated. Then, the polynomial surrogate was used to predict the aerodynamic coefficients and the flutter velocity based on the dataset obtained from the unsteady Reynolds-average Navier–Stokes computational fluid dynamics simulations. The accuracy of the surrogate model is evaluated using error metrics, such as the sum-squared error, R-squared, and the root mean squared error. The results obtained from the statistical metrics demonstrate that the proposed polynomial surrogate provides an accurate prediction of the force coefficients and their derivatives, as well as the critical flutter velocity. A benchmark example is used to demonstrate the efficiency of the proposed model. Finally, a genetic algorithm was introduced to determine an optimal shape, and the results demonstrate that the proposed optimization framework can improve the design process remarkably. [ABSTRACT FROM AUTHOR]

The use of superabsorbent polymers, sometimes known as SAP, is a tremendously efficacious method for reducing the amount of autogenous shrinkage (AS) that occurs in high-performance concrete. This study utilizes support vector regression (SVR) as a standalone machine-learning algorithm (MLA) which is then ensemble with boosting and bagging approaches to reduce the bias and overfitting issues. In addition, these ensemble methods are optimized with twenty sub-models with varying the nth estimators to achieve a robust R2. Moreover, modified bagging as random forest regression (RFR) is also employed to predict the AS of concrete containing supplementary cementitious materials (SCMs) and SAP. The data for modeling of AS includes water to cement ratio (W/C), water to binder ratio (W/B), cement, silica fume, fly ash, slag, the filer, metakaolin, super absorbent polymer, superplasticizer, super absorbent polymer size, curing time, and super absorbent polymer water intake. Statistical and k-fold validation is used to verify the validation of the data using MAE and RMSE. Furthermore, SHAPLEY analysis is performed on the variables to show the influential parameters. The SVM with AdaBoost and modified bagging (RF) illustrates strong models by delivering R2 of approximately 0.95 and 0.98, respectively, as compared to individual SVR models. An enhancement of 67% and 63% in the RF model, while in the case of SVR with AdaBoost, it was 47% and 36%, in RMSE and MAE of both models, respectively, when compared with the standalone SVR model. Thus, the impact of a strong learner can upsurge the efficiency of the model. [ABSTRACT FROM AUTHOR]

This study aims to solve the non-linear fifth-order induction motor model (FO-IMM) using the Gudermannian neural networks (GNN) along with the optimization procedures of global search as a genetic algorithm together with the quick local search process as active-set technique (GNN-GA-AST). The GNN are executed to discretize the non-linear FO-IMM to prompt the fitness function in the procedure of mean square error. The exactness of the GNN-GA-AST is observed by comparing the obtained results with the reference results. The numerical performances of the stochastic GNN-GA-AST are provided to tackle three different variants based on the non-linear FO-IMM to authenticate the consistency, significance and efficacy of the designed stochastic GNN-GA-AST. Additionally, statistical illustrations are available to authenticate the precision, accuracy and convergence of the designed stochastic GNN-GA-AST. [ABSTRACT FROM AUTHOR]

A trustworthy assessment of soil moisture content plays a significant role in irrigation planning and in controlling various natural disasters such as floods, landslides, and droughts. Various machine learning models (MLMs) have been used to increase the accuracy of soil moisture content prediction. The present investigation aims to apply MLMs with novel structures for the estimation of daily volumetric soil water content, based on the stacking of the multilayer perceptron (MLP), random forest (RF), and support vector regression (SVR). Two groups of input variables were considered: the first (Model A) consisted of various meteorological variables (i.e., daily precipitation, air temperature, humidity, and wind speed), and the second (Model B) included only daily precipitation. The stacked model (SM) had the best performance (R2 = 0.962) in the prediction of daily volumetric soil water content for both categories of input variables when compared with the MLP (R2 = 0.957), RF (R2 = 0.956) and SVR (R2 = 0.951) models. Overall, the SM, which, in general, allows the weaknesses of the individual basic algorithms to be overcome while still maintaining a limited number of parameters and short calculation times, can lead to more accurate predictions of soil water content than those provided by more commonly employed MLMs.

A hierarchical method of mathematical and computer modeling of interval-stochastic thermal processes in complex electronic systems for various purposes is developed. The developed concept of hierarchical structuring reflects both the constructive hierarchy of a complex electronic system and the hierarchy of mathematical models of heat exchange processes. Thermal processes that take into account various physical phenomena in complex electronic systems are described by systems of stochastic, unsteady, and nonlinear partial differential equations and, therefore, their computer simulation encounters considerable computational difficulties even with the use of supercomputers. The hierarchical method avoids these difficulties. The hierarchical structure of the electronic system design, in general, is characterized by five levels: Level 1 - the active elements of the ES (microcircuits, electro-radio-elements); Level 2 - electronic module; Level 3 - a panel that combines a variety of electronic modules; Level 4 - a block of panels; Level 5 - stand installed in a stationary or mobile room. The hierarchy of models and modeling of stochastic thermal processes is constructed in the reverse order of the hierarchical structure of the electronic system design, while the modeling of interval-stochastic thermal processes is carried out by obtaining equations for statistical measures. The hierarchical method developed in the article allows to take into account the principal features of thermal processes, such as the stochastic nature of thermal, electrical and design factors in the production, assembly and installation of electronic systems, stochastic scatter of operating conditions and the environment, non-linear temperature dependencies of heat exchange factors, unsteady nature of thermal processes. The equations obtained in the article for statistical measures of stochastic thermal processes are a system of 14 non-stationary nonlinear differential equations of the first order in ordinary derivatives, whose solution is easily implemented on modern computers by existing numerical methods. The results of applying the method for computer simulation of stochastic thermal processes in electron systems are considered. The hierarchical method is applied in practice for the thermal design of real electronic systems and the creation of modern competitive devices.

This research work is to design a neural-swarming heuristic procedure for numerical investigations of Singular Multi-Pantograph Delay Differential (SMP-DD) equation by applying the function approximation aptitude of Artificial Neural Networks (ANNs) optimized efficient swarming mechanism based on Particle Swarm Optimization (PSO) integrated with convex optimization with Active Set (AS) algorithm for rapid refinements, named as ANN-PSO-AS. A merit function (MF) on mean squared error sense is designed by using the differential ANN models and boundary condition. The optimization of this MF is executed with the global PSO and local search AS approaches. The planned ANN-PSO-AS approach is instigated for three different SMP-DD model-based equations. The assessment with available standard results relieved the effectiveness, robustness and precision that is further authenticated through statistical investigations of Variance Account For, Root Mean Squared Error, Semi-Interquartile Range and Theil's inequality coefficient performances. [ABSTRACT FROM AUTHOR]