Search

Accurately determining the favorable areas of geothermal resources and selecting the target positions of exploration wells are extremely important for exploration and efficient development. This study used the Pearson correlation coefficient and Gini gain to analyze five influencing factors related to the presence of economically viable geothermal potential. The evaluation model of the favorable areas was constructed by using different Machine Learning (ML) methods: Bayesian classifier (Bayes), Support Vector Machine, Bootstrap Aggregating (Bagging), BP neural network, Decision Tree and Logistic Regression classification. The quality of each model was verified by statistical evaluation indicators: Accuracy (ACC), F1 score (F1) and Receiver Operating Characteristic curve (ROC curve). The methodology was applied to the case study of Xinjiang Uygur Autonomous Region, China. Due to the results obtained, all ML models showed strong prediction and classification performance on the target area selection of geothermal exploration, as evidenced by each model’s metrics: the ACC was above 80%, the F1 was above 0.8, and the Area Under the ROC Curve (AUC) was greater than 0.85. The metrics obtained by the Bagging method were the highest. Finally, the results of the six ML models were combined to classify the study area’s geothermal potential, which was consistent with the available information. This study provides a specific basis and technical support for applying the method in further surveys and campaigns. [ABSTRACT FROM AUTHOR]

Objective To solve the difficulty of obtaining a radar cross section (RCS) using traditional simulation and measurement methods under high frequency, this study proposes a hybrid method which combines bootstrap aggregation (Bagging) and spectral mixture covariance function-based Gaussian process regression (GPR) model to predict the RCS of ships in the high frequency band efficiently and accurately according to the data in the low frequency band.MethodsFirst, according to the monostatic RCS data of ships in the low frequency band, the training subset is obtained by resampling. The spectral mixture covariance function-based GPR model is then used to extrapolate the RCS data of each subset in the frequency domain. Finally, the extrapolation results of each subset are mixed by the Bagging method to further improve the extrapolation accuracy and robustness of GPR. The proposed method is then tested on the simulation data and measured data respectively. ResultsThe predicted value of the Bagging-GPR hybrid method is basically consistent with the simulated value and measured value, and the root mean square error is very small.ConclusionsThe Bagging-GPR hybrid method has high RCS extrapolation accuracy and good robustness in the frequency domain, providing a new technical means for quickly obtaining the high-frequency RCS characteristics of targets.

This research work explores the use of single neural networks and bootstrap aggregated neural networks for predicting hourly global solar radiation. A database of 3606 data points were from the Renewable Energies Development Center, radiometric station ‘Shems’ of Bouzareah. The single neural networks and bootstrap aggregated neural networks were built together. The precision and durability of neural network models generated with an incomplete quantity of training datasets were improved using bootstrap aggregated neural networks. To produce numerous sets of training data points, the training data was re-sampled utilising bootstrap resampling by replacement. A neural network model was built for each of the data points. The individual neural network models were then combined to produce the bootstrap aggregated neural networks. The experimental and predicted values of global solar radiation were compared, and lower root mean squared errors (68.3968 and 62.4856 Wh m−2) were discovered during the testing phases for single neural networks and bootstrap aggregated neural networks, respectively. The results of these models show that the bootstrap aggregated neural networks model is more accurate and robust than single neural networks.

This research work explores the use of single neural networks and bootstrap aggregated neural networks for predicting hourly global solar radiation. A database of 3606 data points were from the Renewable Energies Development Center, radiometric station ‘Shems’ of Bouzareah. The single neural networks and bootstrap aggregated neural networks were built together. The precision and durability of neural network models generated with an incomplete quantity of training datasets were improved using bootstrap aggregated neural networks. To produce numerous sets of training data points, the training data was re-sampled utilising bootstrap resampling by replacement. A neural network model was built for each of the data points. The individual neural network models were then combined to produce the bootstrap aggregated neural networks. The experimental and predicted values of global solar radiation were compared, and lower root mean squared errors (68.3968 and 62.4856 Wh m−2) were discovered during the testing phases for single neural networks and bootstrap aggregated neural networks, respectively. The results of these models show that the bootstrap aggregated neural networks model is more accurate and robust than single neural networks. [ABSTRACT FROM AUTHOR]

The intermittent nature of the output power of photovoltaic (PV) systems, in addition to the fast-varying solar irradiance, has prompted the development of fast, accurate, and reliable forecasting techniques. This paper presents a comparative study of five ensemble machine learning methods based on bootstrap aggregating and gradient boosting for PV applications, namely AdaBoost, LightGBM, XGBoost, Random Forest, and CatBoost. A dataset of fast-varying environmental conditions was collected, and the terminal current of the experimental setup was augmented by applying a mathematical model, along with an evolutionary algorithm to extract the parasitic resistances found in the Single Diode Model (SDM) to accommodate for the aging effect. The mathematical model was evaluated for several irradiance and temperature levels against manufacturer Standard Test Conditions (STC), and then variations of environmental conditions were adjusted based on the manufacturer datasheet. CatBoost showed the lowest overall absolute error distribution (with respect to the mean and standard deviation) of all methods, and the best performance in terms of the absolute error (0.25%) and its standard deviation (0.85%) relative to the mathematical model. The AdaBoost method had the highest absolute error (34.5%) and a standard deviation of (15.8%). After hyperparameters tuning, CatBoost still outperformed other methods and showed consistency of high accuracy above 99% in performance with respect to a testing dataset, in addition, to having the largest area under the curve using the trapezoidal rule. Therefore, the CatBoost prediction method is expected to be an effective technique for maximum power point tracking schemes under fast-varying environmental conditions.

Automated plant recognition based on leaf images is a challenging task among the researchers from several fields. This task requires distinguishing features derived from leaf images for assigning class label to a leaf image. There are several methods in literature for extracting such distinguishing features. In this paper, we propose a novel automated framework for leaf identification. The proposed framework works in multiple phases i.e. pre-processing, feature extraction, classification using bagging approach. Initially, leaf images are pre-processed using image processing operations such as boundary extraction and cropping. In the feature extraction phase, popular nature inspired optimization algorithms viz. Spider Monkey Optimization (SMO), Particle Swarm Optimization (PSO) and Gray Wolf Optimization (GWO) have been exploited for reducing the dimensionality of features. In the last phase, a leaf image is classified by multiple classifiers and then output of these classifiers is combined using majority voting. The effectiveness of the proposed framework is established based on the experimental results obtained on three datasets i.e. Flavia, Swedish and self-collected leaf images. On all the datasets, it has been observed that the classification accuracy of the proposed method is better than the individual classifiers. Furthermore, the classification accuracy for the proposed approach is comparable to deep learning based method on the Flavia dataset. [ABSTRACT FROM AUTHOR]

In general, studies on short-term hourly electricity load modeling and forecasting do not investigate in detail the sources of uncertainty in forecasting. This study aims to evaluate the impact and benefits of applying bootstrap aggregation in overcoming the uncertainty in time series forecasting, thereby increasing the accuracy of multistep ahead point forecasts. We implemented the existing and proposed clustering-based bootstrapping methods to generate new electricity load time series. In the proposed method, we use singular spectrum analysis to decompose the series between signal and noise to reduce the variance of the bootstrapped series. The noise is then bootstrapped by K-means clustering-based generation of Gaussian normal distribution (KM.N) before adding it back to the signal, resulting in the bootstrapped series. We apply the benchmark models for electricity load forecasting, SARIMA, NNAR, TBATS, and DSHW, to model all new bootstrapped series and determine the multistep ahead point forecasts. The forecast values obtained from the original series are compared with the mean and median across all forecasts calculated from the bootstrapped series using the Malaysian, Polish, and Indonesian hourly load series for 12, 24, and 36 steps ahead. We conclude that, in this case, the proposed bootstrapping method improves the accuracy of multistep-ahead forecast values, especially when considering the SARIMA and NNAR models. [ABSTRACT FROM AUTHOR]

Sparse functional data are commonly observed in real-data analyzes. For such data, we propose a new classification method based on functional principal component analysis (FPCA) and bootstrap aggregating. Bootstrap aggregating is believed to improve the single classifier. In this paper, we apply this belief to an FPCA based classification, and compare the classification performance with that of the single classifiers. The simulation results show that the proposed method performs better than the conventional single classifiers. We then conduct two real-data analyzes. [ABSTRACT FROM AUTHOR]

This study presents a model of computer-aided intelligence capable of automatically detecting positive COVID-19 instances for use in regular medical applications. The proposed model is based on an Ensemble boosting Neural Network architecture and can automatically detect discriminatory features on chest X-ray images through Two Step-As clustering algorithm with rich filter families, abstraction and weight-sharing properties. In contrast to the generally used transformational learning approach, the proposed model was trained before and after clustering. The compilation procedure divides the datasets samples and categories into numerous sub-samples and subcategories and then assigns new group labels to each new group, with each subject group displayed as a distinct category. The retrieved characteristics discriminant cases were used to feed the Multiple Neural Network method, which was then utilised to classify the instances. The Two Step-AS clustering method has been modified by pre-aggregating the dataset before applying Multiple Neural Network algorithm to detect COVID-19 cases from chest X-ray findings. Models for Multiple Neural Network and Two Step-As clustering algorithms were optimised by utilising Ensemble Bootstrap Aggregating algorithm to reduce the number of hyper parameters they include. The tests were carried out using the COVID-19 public radiology database, and a cross-validation method ensured accuracy. The proposed classifier with an accuracy of 98.02% percent was found to provide the most efficient outcomes possible. The result is a low-cost, quick and reliable intelligence tool for detecting COVID-19 infection. [ABSTRACT FROM AUTHOR]

Population poverty is one of the serious problems in Indonesia. The percentage of population poverty used as a means for a statistical instrument to be guidelines to create standard policies and evaluations to reduce poverty. The aims of the research are to determine model population poverty using MARS and Bagging MARS then to understand the most influence variable population poverty of Central Java Province in 2018. The result of this research is the Bagging MARS model showed better accuracy than the MARS model. Since, GCV value in the Bagging MARS model is 0,009798721 and GCV value in the MARS model is 6,985571. The most influential variable poverty population of Central Java Province in 2018 in the MARS model is the percentage of the old school expectation rate (X9). Then, the most influential variable in the Bagging MARS model is the number of diarrhea (X1).

Lithium-ion batteries are widely employed in industries and daily life. Research on the state of health (SOH) of batteries is essential for grasping the performance of batteries, better guiding battery health management, and avoiding safety mishaps caused by battery aging. Nowadays, most research adopts a data-driven artificial intelligence approach to assess SOH. However, the majority of approaches are based on entire voltage, current, or temperature curves. In reality, voltage, current, and temperature are frequently presented in segments, leading to the limited flexibility and slow analysis speed of the traditional techniques. This study solves the problem by dividing the whole voltage curve into many typical kinds of segments with equal timescales based on different typical voltage beginning points. On this foundation, the temporal convolution network (TCN) is used to create a sub-model of SOH estimation for several typical kinds of segments. In addition, the sub-models are fused using the bootstrap aggregating (Bagging) approach to boost accuracy. Finally, this research uses a publicly available dataset from Oxford to demonstrate the effectiveness of the suggested strategy.

The majority of eXplainable Artificial Intelligence (XAI) methods assume the local linearity of the decision boundary, leading to significant errors when dealing with non-linear local decision boundaries. Moreover, explanation methods relying on perturbation samples yield only a constant value for feature contributions. This study aims to overcome these limitations by introducing a novel explanation method that accurately captures non-linear local classification boundaries or regression splines while also quantifying the uncertainty associated with each feature contribution. Expanding upon the state-of-the-art XAI method, Local Interpretable Model-agnostic Explanations (LIME), we propose a nonlinear local explainer called BMB-LIME (Bootstrap aggregating Multivariate adaptive regression splines Bayesian LIME). BMB-LIME is constructed using weighted multivariate adaptive regression splines (MARS) with bootstrap aggregating and models the uncertainty of feature contributions within a Bayesian framework. Through a series of experiments, we demonstrate the superior performance of the BMB-LIME local explainer over baseline methods in terms of local fidelity and stability. Its adjusted r-squared is statistically significant by at least 80.5 percent, and it consistently demonstrates over 89.1 percent consistency in feature contribution and Jaccard similarity across diverse real-world and simulated datasets. The proposed method not only enhances fidelity and stability but also improves the measurement of uncertainty in explanations, thereby significantly contributing to the trustworthiness and transparency of XAI. [ABSTRACT FROM AUTHOR]

Wavelet thresholding (or shrinkage) attempts to remove the noises existing in the signals while preserving inherent pattern characteristics in the reconstruction of true signals. For data-denoising purpose, we present a new wavelet thresholding procedure which employs the step-down testing idea of identifying active contrasts in unreplicated fractional factorial experiments. The proposed method employs bootstrapping methods to a step-down test for thresholding wavelet coefficients. By introducing the concept of a false discovery error rate in testing wavelet coefficients, we shrink the wavelet coefficients with p-values higher than the error rate. The error rate controls the expected proportion of wrongly accepted coefficients among chosen wavelet coefficients. Bootstrap samples are used to approximate the p-value for computational efficiency. We also present some guidelines for selecting the values of hyper-parameters which affect the performance in the step-down thresholding procedure. Based on some common testing signals and an air-conditioner sounds example, the comparison of our proposed procedure with other thresholding methods in the literature is performed. The analytical results show that the proposed procedure has a potential in data-denoising and data-reduction in a variety of signal reconstruction applications.

In general, studies on short-term hourly electricity load modeling and forecasting do not investigate in detail the sources of uncertainty in forecasting. This study aims to evaluate the impact and benefits of applying bootstrap aggregation in overcoming the uncertainty in time series forecasting, thereby increasing the accuracy of multistep ahead point forecasts. We implemented the existing and proposed clustering-based bootstrapping methods to generate new electricity load time series. In the proposed method, we use singular spectrum analysis to decompose the series between signal and noise to reduce the variance of the bootstrapped series. The noise is then bootstrapped by K-means clustering-based generation of Gaussian normal distribution (KM.N) before adding it back to the signal, resulting in the bootstrapped series. We apply the benchmark models for electricity load forecasting, SARIMA, NNAR, TBATS, and DSHW, to model all new bootstrapped series and determine the multistep ahead point forecasts. The forecast values obtained from the original series are compared with the mean and median across all forecasts calculated from the bootstrapped series using the Malaysian, Polish, and Indonesian hourly load series for 12, 24, and 36 steps ahead. We conclude that, in this case, the proposed bootstrapping method improves the accuracy of multistep-ahead forecast values, especially when considering the SARIMA and NNAR models.

In this era of advancements in deep learning, an autonomous system that recognizes handwritten characters and texts can be eventually integrated with the software to provide better user experience. Like other languages, Bangla handwritten text extraction also has various applications such as post-office automation, signboard recognition, and many more. A large-scale and efficient isolated Bangla handwritten character classifier can be the first building block to create such a system. This study aims to classify the handwritten Bangla characters. The proposed methods of this study are divided into three phases. In the first phase, seven convolutional neural networks i.e., CNN-based architectures are created. After that, the best performing CNN model is identified, and it is used as a feature extractor. Classifiers are then obtained by using shallow machine learning algorithms. In the last phase, five ensemble methods have been used to achieve better performance in the classification task. To systematically assess the outcomes of this study, a comparative analysis of the performances has also been carried out. Among all the methods, the stacked generalization ensemble method has achieved better performance than the other implemented methods. It has obtained accuracy, precision, and recall of 98.68%, 98.69%, and 98.68%, respectively on the Ekush dataset. Moreover, the use of CNN architectures and ensemble methods in large-scale Bangla handwritten character recognition has also been justified by obtaining consistent results on the BanglaLekha-Isolated dataset. Such efficient systems can move the handwritten recognition to the next level so that the handwriting can easily be automated.

Introduction: Observational research in the field of health often does not conduct randomized controlled trials on research subjects. A non-random selection process on research subjects can result in a biased treatment effect due to an imbalance between the treatment and control groups. Methods: The problem of bias effects can be dealt with by reducing the bias in the confounding variable using the propensity score method. Estimation of propensity score can use machine learning method with a classification tree analysis approach. The resulting single classification tree model is still unstable if there is a slight change in learning data. Therefore, the ensemble method is applied which is bootstrap aggregating the classification tree as a tool to improve the stability and predictive power of the classification tree. Results: This study aims to determine the effect of giving treatment antiretroviral therapy and counseling to opportunistic infections in HIV AIDS patients. The result of propensity score stratification analysis using bootstrap aggregating classification trees analysis is able to reduce the bias by 89.54%, using 5 strata and having a balanced covariate in each stratum. Conclusion: Testing the effect of treatment shows that there is a significant effect of giving antiretroviral therapy and counseling to opportunistic infections in HIV AIDS patients.

Ensemble classifiers have been proven to result in better classification accuracy than that of a single strong learner in many machine learning studies. Although many studies on electroencephalography-brain-computer interface (BCI) used ensemble classifiers to enhance the BCI performance, ensemble classifiers have hardly been employed for near-infrared spectroscopy (NIRS)-BCIs. In addition, since there has not been any systematic and comparative study, the efficacy of ensemble classifiers for NIRS-BCIs remains unknown. In this study, four NIRS-BCI datasets were employed to evaluate the efficacy of linear discriminant analysis ensemble classifiers based on the bootstrap aggregating. From the analysis results, significant (or marginally significant) increases in the bitrate as well as the classification accuracy were found for all four NIRS-BCI datasets employed in this study. Moreover, significant bitrate improvements were found in two of the four datasets.

Ensemble classifiers have been proven to result in better classification accuracy than that of a single strong learner in many machine learning studies. Although many studies on electroencephalography-brain-computer interface (BCI) used ensemble classifiers to enhance the BCI performance, ensemble classifiers have hardly been employed for near-infrared spectroscopy (NIRS)-BCIs. In addition, since there has not been any systematic and comparative study, the efficacy of ensemble classifiers for NIRS-BCIs remains unknown. In this study, four NIRS-BCI datasets were employed to evaluate the efficacy of linear discriminant analysis ensemble classifiers based on the bootstrap aggregating. From the analysis results, significant (or marginally significant) increases in the bitrate as well as the classification accuracy were found for all four NIRS-BCI datasets employed in this study. Moreover, significant bitrate improvements were found in two of the four datasets. [ABSTRACT FROM AUTHOR]

The present article proposes a methodology to consider the uncertainty intrinsic to data-based models when comparing their performance. The goal is to provide a quantification of the variability of this type of models due to the random nature of the calibration process and enable a statistical comparison of the models' performance when attempting to identify the best. The methodology proposed doesn't provide an alternative metric to determine the models' performance, but it expands the traditional deterministic comparison to a stochastic comparison. The methodology builds on the current standard approach for developing data-based model and its application is demonstrated to model sewer condition using data from 4 trunk sewers of the SANEST – Saneamento da Costa do Estoril sewer system, corresponding to 25 km of sewer pipes. The data-based models were developed using artificial neural networks, support vector machines, bootstrapping aggregation and least squares support vector machines. For the case study, the highest and average misclassification performance records are similar for all models (23% to 24% and 31% to 33%, respectively) but the lowest performance varied more significantly (39% to 62%). This demonstrates that selecting a model based on its maximum single realisation performance alone may be misleading. [ABSTRACT FROM AUTHOR]

• Bagging and DT models produce reliable predictions for the prediction of the compressive strength of geopolymer concrete. • LSBoost model can perform better than Bagging, DT and GEP models. • Knowledge gaps and future recommendations are identified. The compressive strength (f c) of the concrete is an important parameter in the structural design. However, the assessment of f c via an experimental program is time-consuming, costly, and needs a labor force. Therefore, the forecasting of f c through different algorithms can accelerate and facilitate this process and also provide guidance for scheduling the progress of the construction. While some studies have explored the use of models for the prediction of f c of concrete, the ensemble models that can predict the f c of GPC with industrial by-products is still lacking. Within this scope, decision tree (DT), Bootstrap aggregating (Bagging), and Least-squares boosting (LSBoost) models were devised to predict f c of ground granulated blast furnace slag (GGBFS)-based geopolymer concrete (GPC). The data points collected to devise a GEP model in the previous study were used and the prediction results of the GEP model were compared with the proposed ensemble models in the current study. The age of the specimen, NaOH solution concentration, natural zeolite (NZ) content, silica fume (SF) content, and GGBFS content were used as input parameters, and f c was used as output parameter. According to ANOVA analysis, the age of the specimen was found as the most influential parameter in the determination of the f c of GGBFS-based GPC. Also, Multiple linear regression equation was proposed to estimate the f c of GGBFS-based GPC with the accuracy of 93%. The most accurate model was introduced through performance metrics and the Taylor diagram. The results proved that the highest accuracy and stable predictions were achieved by the LSBoost model with R-squared value of 98.25% followed by GEP model developed in the previous study, DT and Bagging models. However, it is worth mentioning that due to having a high coefficient of correlation values (>%80), DT and Bagging models also have an acceptable ability for predicting f c of GGBS-based GPC. [ABSTRACT FROM AUTHOR]

In semiconductor manufacturing processes, spatial defect patterns on semiconductor wafers can provide useful information to quality engineers regarding the root causes of abnormalities. As early detection of process problems increases the wafer yield, the automatic recognition of defect patterns is crucial. It its intrinsic assumption that a decluttering of noise defects helps in accurately classifying defect patterns. However, most existing noise detection methods require the pre-determination of parameters that significantly affect the detection performance. In this paper, we propose a parameter-free detection framework called BAND, which refers to BAgging Noise Detectors, to solve this problem. By adopting the framework of bootstrap aggregating (bagging), which combines the detection results from multiple outlier detection algorithms, the proposed framework ensures the diversity of randomly chosen parameter values while maintaining an accurate detection of noise defects. Based on the estimated distributions of outlierness scores, a group of outliers is discriminated from the other group of inliers via spectral clustering. This procedure enables the proposed framework to be parameter-free. Moreover, numerical experiments based on synthetic and real-world datasets are presented to demonstrate the effectiveness of the proposed framework. • A novel outlier detection framework for wafer denoising was proposed. • The proposed framework presents the advantage of being parameter-free. • The proposed framework shows superior detection performances over existing methods. • Semiconductor engineers can easily use the proposed framework for wafer inspection. [ABSTRACT FROM AUTHOR]

Partial discharge (PD) is regarded as a precursor to plant failure and therefore, an effective indication of plant condition. Locating the source of PD before failure is key to efficient maintenance and improving reliability of power systems. This paper presents a low cost, autonomous partial discharge radiolocation mechanism to improve PD localization precision. The proposed radio frequency-based technique uses the wavelet packet transform (WPT) and machine learning ensemble methods to locate PDs. More specifically, the received signals are decomposed by the WPT and analyzed in order to identify localized PD signal patterns in the presence of noise. The regression tree algorithm, bootstrap aggregating method, and regression random forest are used to develop PD localization models based on the WPT-based PD features. The proposed PD localization scheme has been found to successfully locate PD with negligible error. Additionally, the principle of the PD location scheme has been validated using a separate test dataset. Numerical results demonstrate that the WPT-random forest PD localization scheme produced superior performance as a result of its robustness against noise. [ABSTRACT FROM AUTHOR]

Recognition of medieval handwritten Gurmukhi manuscripts is an essential process for resourceful contents exploitation of the priceless information contained in them. There are numerous Gurmukhi script ancient manuscripts from fifteenth to twentieth century's. In this paper, we have considered, work written by various persons from 18th to 20th centuries. For recognition, we have used various feature extraction techniques like zoning, discrete cosine transformations, and gradient features and different combinations of these features. For classification, four classifiers, namely, k-NN, SVM, Decision Tree, Random Forest individual and combinations of these four classifiers with voting scheme have been considered. Adaptive boosting and bagging have been explored for improving the recognition results and achieves the new state of the art for recognition of medieval handwritten Gurmukhi manuscripts recognition. Using this proposed framework, maximum recognition accuracy of 95.91% has been achieved using adaptive boosting technique and a combination of four different classifiers considered in this paper. To the best of our knowledge, this work is the successful attempt towards recognition of medieval handwritten Gurmukhi manuscripts and it can lead towards the development of optical character recognition systems for recognizing medieval handwritten documents in other Indic and non-Indic scripts as well. [ABSTRACT FROM AUTHOR]

An ENsemble Deep Learning Optimal Treatment (EndLot) approach is proposed for personalized medicine problems. The statistical framework of the proposed method is based on the outcome weighted learning (OWL) framework which transforms the optimal decision rule problem into a weighted classification problem. We further employ an ensemble of deep neural networks (DNNs) to learn the optimal decision rule. Utilizing the flexibility of DNNs and the stability of bootstrap aggregation, the proposed method achieves a considerable improvement over existing methods. An R package "ITRlearn" is developed to implement the proposed method. Numerical performance is demonstrated via simulation studies and a real data analysis of the Cancer Cell Line Encyclopedia data. [ABSTRACT FROM AUTHOR]

To better assess the relationships between environmental exposures and health outcomes, an appropriate spatiotemporal interpolation is critical. Traditional spatiotemporal interpolation methods either consider the spatial and temporal dimensions separately or incorporate both dimensions simultaneously by simply treating time as another dimension in space. Such interpolation results suffer from relatively low accuracy as the true space-time domain is skewed inappropriately and the distance calculation in such domain is not accurate. We employ the efficient k-d tree structure to store spatiotemporal data and adopt several machine learning methods to learn optimal parameters. To overcome the computational difficulty with large data sets, we implement our method on an efficient cluster computing framework - Apache Spark. Real world PM2.5 data sets are utilized to test our implementation and the experimental results demonstrate the computational power of our method, which significantly outperforms the previous work in terms of both speed and accuracy. [ABSTRACT FROM AUTHOR]

Groundwater vulnerability assessment is a measure of potential groundwater contamination for areas of interest. The main objective of this study is to modify original DRASTIC model using four objective methods, Weights-of-Evidence (WOE), Shannon Entropy (SE), Logistic Model Tree (LMT), and Bootstrap Aggregating (BA) to create a map of groundwater vulnerability for the Sari-Behshahr plain, Iran. The study also investigated impact of addition of eight additional factors (distance to fault, fault density, distance to river, river density, land-use, soil order, geological time scale, and altitude) to improve groundwater vulnerability assessment. A total of 109 nitrate concentration data points were used for modeling and validation purposes. The efficacy of the four methods was evaluated quantitatively using the Area Under the Receiver Operating Characteristic (ROC) Curve (AUC). AUC value for original DRASTIC model without any modification of weights and rates was 0.50. Modification of weights and rates resulted in better performance with AUC values of 0.64, 0.65, 0.75, and 0.81 for BA, SE, LMT, and WOE methods, respectively. This indicates that performance of WOE is the best in assessing groundwater vulnerability for DRASTIC model with 7 factors. The results also show more improvement in predictability of the WOE model by introducing 8 additional factors to the DRASTIC as AUC value increased to 0.91. The most effective contributing factor for ground water vulnerability in the study area is the net recharge. The least effective factors are the impact of vadose zone and hydraulic conductivity. [ABSTRACT FROM AUTHOR]

This research was conducted to determine the variables that significantly influence nutritional status of children based on indicators that defined as height for age (H/A) and to classify children nutritional status into normal, short or very short categories. Height for age (H/A) is indicator used to describe the circumstances of malnutrition short. Short children (stunting) is children who fail to reach optimal growth. The secondary data was list of 116 data of children aged 24-59 months at UPT. Puskesmas Klungkung I in 2015. The method was used was ordinal logistic regression and bagging ordinal logistic regression. Based on the research results, it was obtained variables children body length at birth, birth weight, and length of mid-upper arm circumference (MUAC) in pregnant woman were significantly affects the nutritional status of children by the classification accuracy level of ordinal logistic regression and misclassification . Classification accuracy of ordinal logistic regression can be improved by bagging ordinal logistic regression method. Bagging works well on classification method which has unstable procedures. One of classification method which has unstable procedures is ordinal logistic regression. Bagging ordinal logistic regression method by 501 times replication capable to improve classification accuracy of ordinal logistic regression model from to , increased .

Presently, due to the extended availability of gigantic information networks and the beneficial application of graph analysis in various scientific fields, the necessity for efficient and highly scalable community detection algorithms has never been more essential. Despite the significant amount of published research, the existing methods—such as the Girvan–Newman, random-walk edge betweenness, vertex centrality, InfoMap, spectral clustering, etc.—have virtually been proven incapable of handling real-life social graphs due to the intrinsic computational restrictions that lead to mediocre performance and poor scalability. The purpose of this article is to introduce a novel, distributed community detection methodology which in accordance with the community prediction concept, leverages the reduced complexity and the decreased variance of the bagging ensemble methods, to unveil the subjacent community hierarchy. The proposed approach has been thoroughly tested, meticulously compared against different classic community detection algorithms, and practically proven exceptionally scalable, eminently efficient, and promisingly accurate in unfolding the underlying community structure.

Difficulties are faced when formulating hydrological processes, including that of evapotranspiration (ET). Conventional empirical methods for formulating these possess some shortcomings. The artificial intelligence approach emerges as the best possible solution to map the relationships between climatic parameters and ET, even with limited knowledge of the interactions between variables. This review presents the state-of-the-art application of artificial intelligence models in ET estimation, along with different types and sources of data. This paper discovers the most significant climatic parameters for different climate patterns. The characteristics of the basic artificial intelligence models are also explored in this review. To overcome the pitfalls of the individual models, hybrid models which use techniques such as data fusion and ensemble modeling, data decomposition as well as remote sensing-based hybridization, are introduced. In particular, the principles and applications of the hybridization techniques, as well as their combinations with basic models, are explained. The review covers most of the related and excellent papers published from 2011 to 2019 to keep its relevancy in terms of time frame and field of study. Guidelines for the future prospects of ET estimation in research are advocated. It is anticipated that such work could contribute to the development of agriculture-based economy.

Olive cultivation is of great importance in Southern Europe but olive pollen is the leading cause of allergy in many regions where it is grown. The best preventive measure for allergic patients is to avoid exposure. Thus, aerobiological monitoring networks must supply realistic pollen classes for the different types of allergic pollen. Even though those pollen classes are defined, they do not necessarily fit local data. Altogether, they should use predictive models to assess flowering intensity in advance. In this study, the Olea pollen degree of exposure classes (OPDEC) are defined based on percentiles and a predictive model is suggested for Cartagena, Spain. 24 year (1993–2016) Olea pollen counts series was used to characterize the Main Pollen Season (MPS). The aerobiological samples were processed following the methodology proposed by Hirst and developed by the Spanish Aerobiology Network. The aerobiological database was completed with the meteorological data supplied by AEMET (Spanish State Meteorological Agency). MPS evolution over time, and its relation with temperature and rainfall, has been analysed. The study showed an increase in MPS duration and the amount of Olea pollen grains collected both in MPS and the peak day. The OPDEC should fit local data to improve preventive measures. Based on the 24 year series, the proposed OPDEC for Cartagena are: Low (≤ 10 grains/m 3 ), Medium (between 10 and 50 grains/m 3 ), High (between 51 and 100 grains/m 3 ) and Very High (≥ 100 grains/m 3 ). Olea pollen estimations in the MPS and in the peak day were obtained by means of three Regression Methods and climatic factors. The analysis reveals that the Bagging for Regression Trees (BRT) method is a good predictive alternative and stablishes the importance for each meteorological variable. [ABSTRACT FROM AUTHOR]

Recently, the Entropy Ensemble Filter (EEF) method was proposed to mitigate the computational cost of the Bootstrap AGGregatING (bagging) method. This method uses the most informative training data sets in the model ensemble rather than all ensemble members created by the conventional bagging. In this study, we evaluate, for the first time, the application of the EEF method in Neural Network (NN) modeling of El Nino-southern oscillation. Specifically, we forecast the first five principal components (PCs) of sea surface temperature monthly anomaly fields over tropical Pacific, at different lead times (from 3 to 15 months, with a three-month increment) for the period 1979-2017. We apply the EEF method in a multiple-linear regression (MLR) model and two NN models, one using Bayesian regularization and one Levenberg-Marquardt algorithm for training, and evaluate their performance and computational efficiency relative to the same models with conventional bagging. All models perform equally well at the lead time of 3 and 6 months, while at higher lead times, the MLR model's skill deteriorates faster than the nonlinear models. The neural network models with both bagging methods produce equally successful forecasts with the same computational efficiency. It remains to be shown whether this finding is sensitive to the dataset size. [ABSTRACT FROM AUTHOR]

Over the past two decades, the Bootstrap AGGregatING (bagging) method has been widely used for improving simulation. The computational cost of this method scales with the size of the ensemble, but excessively reducing the ensemble size comes at the cost of reduced predictive performance. The novel procedure proposed in this study is the Entropy Ensemble Filter (EEF), which uses the most informative training data sets in the ensemble rather than all ensemble members created by the bagging method. The results of this study indicate efficiency of the proposed method in application to synthetic data simulation on a sinusoidal signal, a sawtooth signal, and a composite signal. The EEF method can reduce the computational time of simulation by around 50% on average while maintaining predictive performance at the same level of the conventional method, where all of the ensemble models are used for simulation. The analysis of the error gradient (root mean square error of ensemble averages) shows that using the 40% most informative ensemble members of the set initially defined by the user appears to be most effective. [ABSTRACT FROM AUTHOR]

Background & Aim: In many medical studies, one data set is used to construct the model, and to test its performance. This approach is prone to over optimization, and leads to statistics with low chance of external validity. Data splitting can be used to create training and test sets but the cost is reduction in power. The aim of this study was to demonstrate the ability of bootstrap aggregating (bagging) in improving performance of classification and regression tree (CART) models. Methods & Materials: CART was applied on a sample of 404 subjects, to identify the factors that encourage people to change their body shape by cosmetic surgeries. Comparing known status of subjects with predicted group, sensitivity and specificity of models were compared. Firstly, all data was used to construct the tree and to test its performance. Secondly, model was fitted on half of data and tested on the second half. Thirdly, bagging was applied in which we drew 100 bootstrap samples. Using each bootstrap data, a tree was constr cted and its performance was tested on the unselected subjects. Final group prediction for each subject was determined following majority voting. Results: When the whole data was used the overall accuracy was 59%. In the test data set and bagging, accuracy reduced to 53% and 56%. Corresponding figures in terms of sensitivity were 60%, 52%, and 55%, respectively. Conclusion: Bagging corrected performance estimates for over optimization. Bagging method produces statistics which has higher chance for external validity.