Resampling Approach Research Articles

Spatial Patterns and Grave Goods Differences at the Cemetery of Vedrovice (Czech Republic): A Resampling Approach to Identity Markers in the Early Neolithic

The spatial patterns and grave good associations at the cemetery of Vedrovice – Široká u Lesa (Czech Republic) are explored using computationally intensive formal methods. Resampling approaches to both intra-site spatial arrangement and the distribution of grave goods are explored in relation to layers of identity represented by different buried body attributes – sex, age, and geographical origin. Polished stone adzes, grinding tools and ceramic bowls are non-randomly distributed grave goods. The investigation of differences among the non-randomly distributed artifacts and attributes of burials indicates that the most significant are dichotomies between sexes, accentuated by both origin and age. Spatial analysis shows clear trends toward clustering between burials, and the emerging clusters are further explored for similarities in burial attributes.

Majority re-sampling via sub-class clustering for imbalanced datasets

ABSTRACT Many real world domain problem datasets are class imbalanced where the number of data in a given class is much less than in the other classes. In related literatures, under- and over-sampling techniques are widely used techniques to re-balance the class imbalanced datasets. However, their limitations include the risk of removing representative majority class data samples and the overfitting problem because of generating a large number of synthetic minority class data samples. Therefore, a novel approach, namely Majority Re-sampling visa Sub-class Clustering (MRSC) is introduced. It uses a clustering algorithm to group the majority class data into several clusters, i.e. sub-classes. Then, a new training set containing multiple sub-classes and a minority class is produced, after which the classifier is trained using this new multi-class dataset which has a lower imbalance ratio than the original dataset. The experimental results obtained using 44 two-class imbalanced datasets show that MRSC combined with the k-NN classifiers, including single and ensemble classifiers, significantly outperforms the other classifiers as well as seven state-of-the-art re-sampling approaches. Moreover, for the clustering algorithms based on affinity propagation and k-means, very similar results can be produced, without significant differences in performance, which indicate the stability of MRSC.

SensMap R package and SensMapGUI shiny web application for sensory and consumer data mapping: Variations on external preference mapping and stability assessment

AbstractThis work addresses the theory of External Preference Mapping technique (EPM) that reliably gives a mapping of human perceptions' ratings considered as subjective data onto objective descriptions generally obtained from measurable analytic data. An R package named SensMap is made available to facilitate the adoption of the EPM technique in different research fields and to propose, in addition to classical features, advanced strategies for map visualization and its stability enhancement. The novelty of SensMap is to provide options in sensory space dimension reduction methods as well as options in consumers' likings prediction models based on sensory descriptors. The fitting quality of these models can be evaluated using several information criteria. Moreover, SensMap provides a methodology to quantify and assess the stability of preference maps from different employed strategies using a computational algorithm that consists in calculations of distances between maps based on resampling approach. A small part in this paper was devoted to evaluate the efficiency of the implemented stability approach by a simulation study and by application to real data from food science field. To make easier the conduction of the multiple types of analysis without requiring high programming knowledge, a very intuitive Graphical User Interface named SensMapGUI is implemented on the SensMap package. It is an R‐developed shiny web‐tool that produces results and graphics quickly and efficiently. The tool is made available at the following address https://cran.r-project.org/web/packages/SensMap/index.html and is dedicated for all sensory practitioners and can be widely used by researchers who need subjective to objective data mapping.Practical ApplicationsResearchers and practitioners frequently require to understand human behaviors by measuring their perceptions and searching for the factors that drive their opinions. The SensMap R package is made available to provide a data visualization map, to allow different analysis strategies and to make available a selection process of the best mapping. The tool can be used through the free and open‐access R software by all sensory scientists and practitioners. SensMap will be regularly updated to allow for changes in dependent R packages and will continue to add new features about mapping. The novelty of the package consists in his incorporation of a nicely designed and easily accessible graphical interface named SensMapGUI that hides all technical complexities for the non‐R users. It is a web‐tool once implemented on line can be used on any device like tablets, iOS applications and Android without need to tedious coding. The tool can also be used by all researchers who need to establish a relationship between subjective and objective data. It can serve in some many investigations such as the evaluation of human state of stress by analogy with their electro‐dermal activity measurements, the explanation of old persons hearing ratings toward road noise on the basis of traffic sound measurements data, the evaluation of consumers' preferences toward some products in the market, etc.

HMDE‐FS: A homogeneous distributed ensemble feature selection framework based on resampling with/without replacement

HMDE‐FS: A homogeneous distributed ensemble feature selection framework based on resampling with/without replacement

A resampling approach for confidence intervals in linear time-series models after model selection

Linear models are commonly used in econometrics to describe structural relations and draw inferences about underlying parameters. Although the theoretical and practical value of these models has been justified under ideal circumstances, empirical studies have shown that they did not well fit the observed data in reality, and their selection is therefore an important problem. In the big-data era, one approach is to consider a large set of covariates and apply model selection. However, many model selection methods can provide consistent coefficient estimators and confidence intervals, but perform poorly in practice due to an insufficient sample size. Especially for financial data, which have relatively small sample sizes and are usually correlated, classical methods to construct confidence intervals after model selection are usually inaccurate. We consider the problem of constructing confidence intervals for selected coefficients in linear time-series models. To overcome the difficulties of post-selection confidence intervals, we introduce consistent estimators of the selected parameters and work on a resampling approach for confidence interval construction. Theory, simulations, and empirical studies demonstrate the advantages of the proposed method.

Exploiting the Generative Adversarial Network Approach to Create a Synthetic Topography Corneal Image.

Corneal diseases are the most common eye disorders. Deep learning techniques are used to perform automated diagnoses of cornea. Deep learning networks require large-scale annotated datasets, which is conceded as a weakness of deep learning. In this work, a method for synthesizing medical images using conditional generative adversarial networks (CGANs), is presented. It also illustrates how produced medical images may be utilized to enrich medical data, improve clinical decisions, and boost the performance of the conventional neural network (CNN) for medical image diagnosis. The study includes using corneal topography captured using a Pentacam device from patients with corneal diseases. The dataset contained 3448 different corneal images. Furthermore, it shows how an unbalanced dataset affects the performance of classifiers, where the data are balanced using the resampling approach. Finally, the results obtained from CNN networks trained on the balanced dataset are compared to those obtained from CNN networks trained on the imbalanced dataset. For performance, the system estimated the diagnosis accuracy, precision, and F1-score metrics. Lastly, some generated images were shown to an expert for evaluation and to see how well experts could identify the type of image and its condition. The expert recognized the image as useful for medical diagnosis and for determining the severity class according to the shape and values, by generating images based on real cases that could be used as new different stages of illness between healthy and unhealthy patients.

RBSP-Boosting: A Shapley value-based resampling approach for imbalanced data classification

Addressing the problem of imbalanced data category distribution in real applications and the problem of traditional classifiers tending to ensure the accuracy of the majority class while ignoring the accuracy of the minority class when processing imbalanced data, this paper proposes a method called RBSP-Boosting for imbalanced data classification. First, RBSP-Boosting introduces the Shapley value and calculates the Shapley value for each sample of the dataset through the truncated Monte Carlo method. Moreover, the proposed method removes the noise data according to the Shapley value and undersamples the samples with Shapley values less than zero in the majority class. Then, it takes the Shapley value as the weight of the sample and oversamples the minority class according to the weight. Finally, the new dataset is trained on the classifier through the AdaBoost classifier. Experiments are conducted on nine groups of UCI and KEEL datasets, and RBSP-Boosting is compared with four sampling algorithms: Random-OverSampler, SMOTE, Borderline-SMOTE and SVM-SMOTE. Experimental results show that the RBSP-Boosting method in the three evaluation metrics of AUC, F-score and G-mean, compared with the best performance of the four comparison algorithms, increases by 4.69%, 10.3% and 7.86%, respectively. The proposed method can significantly improve the effect of imbalanced data classification.

Representative random sampling: an empirical evaluation of a novel bin stratification method for model performance estimation

High-dimensional cancer data can be burdensome to analyze, with complex relationships between molecular measurements, clinical diagnostics, and treatment outcomes. Data-driven computational approaches may be key to identifying relationships with potential clinical or research use. To this end, reliable comparison of feature engineering approaches in their ability to support machine learning survival modeling is crucial. With the limited number of cases often present in multi-omics datasets (“big p, little n,” or many features, few subjects), a resampling approach such as cross validation (CV) would provide robust model performance estimates at the cost of flexibility in intermediate assessments and exploration in feature engineering approaches. A holdout (HO) estimation approach, however, would permit this flexibility at the expense of reliability. To provide more reliable HO-based model performance estimates, we propose a novel sampling procedure: representative random sampling (RRS). RRS is a special case of continuous bin stratification which minimizes significant relationships between random HO groupings (or CV folds) and a continuous outcome. Monte Carlo simulations used to evaluate RRS on synthetic molecular data indicated that RRS-based HO (RRHO) yields statistically significant reductions in error and bias when compared with standard HO. Similarly, more consistent reductions are observed with RRS-based CV. While resampling approaches are the ideal choice for performance estimation with limited data, RRHO can enable more reliable exploratory feature engineering than standard HO.

Photometric redshift uncertainties in weak gravitational lensing shear analysis: models and marginalization

ABSTRACT Recovering credible cosmological parameter constraints in a weak lensing shear analysis requires an accurate model that can be used to marginalize over nuisance parameters describing potential sources of systematic uncertainty, such as the uncertainties on the sample redshift distribution n(z). Due to the challenge of running Markov chain Monte Carlo (MCMC) in the high-dimensional parameter spaces in which the n(z) uncertainties may be parametrized, it is common practice to simplify the n(z) parametrization or combine MCMC chains that each have a fixed n(z) resampled from the n(z) uncertainties. In this work, we propose a statistically principled Bayesian resampling approach for marginalizing over the n(z) uncertainty using multiple MCMC chains. We self-consistently compare the new method to existing ones from the literature in the context of a forecasted cosmic shear analysis for the HSC three-year shape catalogue, and find that these methods recover statistically consistent error bars for the cosmological parameter constraints for predicted HSC three-year analysis, implying that using the most computationally efficient of the approaches is appropriate. However, we find that for data sets with the constraining power of the full HSC survey data set (and, by implication, those upcoming surveys with even tighter constraints), the choice of method for marginalizing over n(z) uncertainty among the several methods from the literature may modify the 1σ uncertainties on Ωm–S8 constraints by ∼4 per cent, and a careful model selection is needed to ensure credible parameter intervals.

Nonparametric inference on smoothed quantile regression process

This paper studies the global estimation in semiparametric quantile regression models. For estimating unknown functional parameters, an integrated quantile regression loss function with penalization is proposed. The first step is to obtain a vector-valued functional Bahadur representation of the resulting estimators, and then derive the asymptotic distribution of the proposed infinite-dimensional estimators. Furthermore, a resampling approach that generalizes the minimand perturbing technique is adopted to construct confidence intervals and to conduct hypothesis testing. Extensive simulation studies demonstrate the effectiveness of the proposed method, and applications to the real estate dataset and world happiness report data are provided.

Predicting Soil Textural Classes Using Random Forest Models: Learning from Imbalanced Dataset

Soil provides a key interface between the atmosphere and the lithosphere and plays an important role in food production, ecosystem services, and biodiversity. Recently, demands for applying machine learning (ML) methods to improve the knowledge and understanding of soil behavior have increased. While real-world datasets are inherently imbalanced, ML models overestimate the majority classes and underestimate the minority ones. The aim of this study was to investigate the effects of imbalance in training data on the performance of a random forest model (RF). The original dataset (imbalanced) included 6100 soil texture data from the surface layer of agricultural fields in northern Iran. A synthetic resampling approach using the synthetic minority oversampling technique (SMOTE) was employed to make a balanced dataset from the original data. Bioclimatic and remotely sensed data, distance, and terrain attributes were used as environmental covariates to model and map soil textural classes. Results showed that based on mean minimal depth (MMD), when imbalanced data was used, distance and annual mean precipitation were important, but when balanced data were employed, terrain attributes and remotely sensed data played a key role in predicting soil texture. Balanced data also improved the accuracies from 44% to 59% and 0.30 to 0.52 with regard to the overall accuracy and kappa values, respectively. Similar increasing trends were observed for the recall and F-scores. It is concluded that, in modeling soil texture classes using RF models through a digital soil mapping approach, data should be balanced before modeling.

A novel nomogram for predicting respiratory adverse events during transport after interventional cardiac catheterization in children.

The rate and predictors of respiratory adverse events (RAEs) during transport discharged from operating room after interventional cardiac catheterization in children remain unclear. This study aimed to investigate the incidence and predictors, and to construct a nomogram for predicting RAEs during transport in this pediatric surgical treatment. This prospective cohort study enrolled 290 consecutive pediatric patients who underwent ventricular septal defects (VSD), atrial septal defects (ASD), and patent ductus arteriosus (PDA) between February 2019 and December 2020. Independent predictors were used to develop a nomogram, and a bootstrap resampling approach was used to conduct internal validation. Composite RAEs were defined as the occurrence of at least 1 complication regarding laryngospasm, bronchospasm, apnea, severe cough, airway secretions, airway obstruction, and oxygen desaturation. The rate of RAEs during transport was 23.1% (67 out of 290). Multivariate analysis identified age (vs. ≤3 years, adjusted odds ratio (aOR) = 0.507, 95% confidence interval (CI), 0.268-0.958, P = 0.036), preoperative upper respiratory tract infections (URI, aOR = 2.335, 95% CI, 1.223-4.460, P = 0.01), type of surgery (vs. VSD, for ASD, aOR = 2.856, 95% CI, 1.272-6.411, P = 0.011; for PDA, aOR = 5.518, 95% CI, 2.425-12.553, P < 0.001), morphine equivalent (vs. ≤0.153 mg/kg, aOR = 2.904, 95% CI, 1.371-6.150, P = 0.005), atropine usage (aOR = 0.463, 95% CI, 0.244-0.879, P = 0.019), and RAEs during extubation to transport (aOR = 5.004, 95% CI, 2.633-9.511, P < 0.001) as independent predictors of RAEs during transport. These six candidate predictors were used to develop a nomogram, which showed a C-statistic value of 0.809 and good calibration (P = 0.844). Internal validation revealed similarly good discrimination (C-statistic, 0.782; 95% CI, 0.726-0.837) and calibration. Decision curve analysis (DCA) also demonstrated the clinical usefulness of the nomogram. The high rate of RAEs during transport reminds us of the need for more medical care and attention. The proposed nomogram can reliably identify pediatric patients at high risk of RAEs during transport and guide clinicians to make proper transport plans. Our findings have important and meaningful implications for RAEs risk prediction, clinical intervention and healthcare quality control.

Longer study length, standardized sampling techniques, and broader geographic scope leads to higher likelihood of detecting stable abundance patterns in long term black-legged tick studies.

BackgroundUnderstanding how study design and monitoring strategies shape inference within, and synthesis across, studies is critical across biological disciplines. Many biological and field studies are short term and limited in scope. Monitoring studies are critical for informing public health about potential vectors of concern, such as Ixodes scapularis (black-legged ticks). Black-legged ticks are a taxon of ecological and human health concern due to their status as primary vectors of Borrelia burgdorferi, the bacteria that transmits Lyme disease. However, variation in black-legged tick monitoring, and gaps in data, are currently considered major barriers to understanding population trends and in turn, predicting Lyme disease risk. To understand how variable methodology in black-legged tick studies may influence which population patterns researchers find, we conducted a data synthesis experiment.Materials and MethodsWe searched for publicly available black-legged tick abundance dataset that had at least 9 years of data, using keywords about ticks in internet search engines, literature databases, data repositories and public health websites. Our analysis included 289 datasets from seven surveys from locations in the US, ranging in length from 9 to 24 years. We used a moving window analysis, a non-random resampling approach, to investigate the temporal stability of black-legged tick population trajectories across the US. We then used t-tests to assess differences in stability time across different study parameters.ResultsAll of our sampled datasets required 4 or more years to reach stability. We also found several study factors can have an impact on the likelihood of a study reaching stability and of data leading to misleading results if the study does not reach stability. Specifically, datasets collected via dragging reached stability significantly faster than data collected via opportunistic sampling. Datasets that sampled larva reached stability significantly later than those that sampled adults or nymphs. Additionally, datasets collected at the broadest spatial scale (county) reached stability fastest.ConclusionWe used 289 datasets from seven long term black-legged tick studies to conduct a non-random data resampling experiment, revealing that sampling design does shape inferences in black-legged tick population trajectories and how many years it takes to find stable patterns. Specifically, our results show the importance of study length, sampling technique, life stage, and geographic scope in understanding black-legged tick populations, in the absence of standardized surveillance methods. Current public health efforts based on existing black-legged tick datasets must take monitoring study parameters into account, to better understand if and how to use monitoring data to inform decisioning. We also advocate that potential future forecasting initiatives consider these parameters when projecting future black-legged tick population trends.

Monte Carlo cross-validation for a study with binary outcome and limited sample size

Cross-validation (CV) is a resampling approach to evaluate machine learning models when sample size is limited. The number of all possible combinations of folds for the training data, known as CV rounds, are often very small in leave-one-out CV. Alternatively, Monte Carlo cross-validation (MCCV) can be performed with a flexible number of simulations when computational resources are feasible for a study with limited sample size. We conduct extensive simulation studies to compare accuracy between MCCV and CV with the same number of simulations for a study with binary outcome (e.g., disease progression or not). Accuracy of MCCV is generally higher than CV although the gain is small. They have similar performance when sample size is large. Meanwhile, MCCV is going to provide reliable performance metrics as the number of simulations increases. Two real examples are used to illustrate the comparison between MCCV and CV.

Benchmarking state-of-the-art imbalanced data learning approaches for credit scoring

The goal of credit scoring is to identify abnormalities, aiding decision making and maintaining the order of financial transactions. Due to the small number of default records, one inevitably faces a class imbalance problem when handling financial data. The class imbalance problem has received a lot of attention because of the economic loss that can occur when one fails to accurately identify default samples. To solve this problem, there are various classic and mature approaches to learning imbalanced data, including resampling approaches, cost-sensitive strategies, and so on. Especially in recent years, generative adversarial networks (GANs) have attracted the attention of researchers to explore these networks’ effects as imbalanced data learning tools. However, no attention has been paid to the systematic scoring and comparison of these traditional and state-of-the-art imbalanced data learning approaches in relation to credit scoring. Therefore, choosing several related datasets, we compare the performance of the traditional approaches and GANs in solving the class imbalance problem of credit scoring; at the same time, with the help of benchmark analysis, we provide some suggestions for relevant research.

Heart failure in obesity: insights from proteomics in patients treated with or without weight-loss surgery

BackgroundObesity is associated with incident heart failure (HF), but the underlying mechanisms are unclear.MethodsWe performed a nested case-control study within the Swedish-Obese-Subjects study, by identifying 411 cases who developed HF and matched them with respect to age, sex, weight-loss-surgery and length of follow-up with 410 controls who did not develop HF. In analyses corrected for multiple testing, we studied 182 plasma proteins known to be related to cardiovascular disease to investigate whether they could add to the understanding of the processes underlying obesity-related HF.ResultsA total of 821 subjects were followed for 16 ± 6 years. Multivariable analysis adjusted for matching variables revealed that 32 proteins were significantly associated with HF. Twelve proteins were related to HF ≥ 80% of the time using a bootstrap resampling approach (false-discovery-rate [FDR] < 0.05): 11 were associated with increased HF-risk: TNFRSF10A*, ST6GAL1, PRCP, MMP12, TIMP1, CCL3, QPCT, ANG, C1QTNF1, SERPINA5 and GAL-9; and one was related to reduced HF-risk: LPL. An further 20 proteins were associated with onset of HF 50–80% of the time using bootstrap resampling (FDR < 0.05). A pathway analysis including all significant 32 proteins suggested that these biomarkers were related to inflammation, matrix remodeling, cardiometabolic hormones and hemostasis. Three proteins, C1QTNF1, FGF-21 and CST3, reflecting dyslipidemia and kidney disease, displayed a higher association with HF in patients who did not undergo weight-loss-surgery and maintained with obesity.ConclusionPathways associated with HF in obesity include inflammation, matrix remodeling, cardiometabolic hormones and hemostasis; three protein biomarkers predicting HF appeared to be obesity-specific.

Development and validation of a prognostic nomogram among patients with acute exacerbation of chronic obstructive pulmonary disease in intensive care unit

BackgroundAcute exacerbation of Chronic Obstructive Pulmonary Disease (AECOPD) contributes significantly to mortality among patients with COPD in Intensive care unit (ICU). This study aimed to develop a nomogram to predict 30-day mortality among AECOPD patients in ICU.MethodsIn this retrospective cohort study, we extracted AECOPD patients from Medical Information Mart for Intensive Care III (MIMIC-III) database. Multivariate logistic regression based on Akaike information criterion (AIC) was used to establish the nomogram. Internal validation was performed by a bootstrap resampling approach with 1000 replications. The discrimination and calibration of the nomogram were evaluated by Harrell’s concordance index (C-index) and Hosmer–Lemeshow (HL) goodness-of-fit test. Decision curve analysis (DCA) was performed to evaluate its clinical application.ResultsA total of 494 patients were finally included in the study with a mean age of 70.8 years old. 417 (84.4%) patients were in the survivor group and 77 (15.6%) patients were in the non-survivor group. Multivariate logistic regression analysis based on AIC included age, pO2, neutrophil-to-lymphocyte ratio (NLR), prognostic nutritional index (PNI), invasive mechanical ventilation and vasopressor use to construct the nomogram. The adjusted C-index was 0.745 (0.712, 0.778) with good calibration (HL test, P = 0.147). The Kaplan–Meier survival curves revealed a significantly lower survival probability in the high-risk group than that in the low-risk group (P < 0.001). DCA showed that nomogram was clinically useful.ConclusionThe nomogram developed in this study could help clinicians to stratify AECOPD patients and provide appropriate care in clinical setting.

Model parameters estimation with non-ignorable missing data using influential exponential tilting resampling approach

This paper proposes to extend [1] mean functional estimation method based on the influential exponential tilting resampling approach (ITRA) to address non-ignorable missing data in linear model parameters statistical inference. The ITRA approach assumes that the nonrespondents’ model corresponds to an exponential tilting of the respondents’ model. The tilted model's specified function is the influential function of the function of interest (parameter). The other basis of the proposed approach is to use the importance resampling techniques to draw inferences about some linear model parameters. Simulation studies were conducted to investigate the performance of the proposed methods and their application to real data. Theoretical justifications are provided as well.

Importance Resampling Approach for Sensor Mutual Coupling Correction

To achieve accurate direction parameter estimation of sensor array, mutual coupling between sensor elements should be properly handled. This problem becomes more challenging when the sources are coherent. In this paper, we derive an importance resampling approach (IRA) for joint mutual coupling and DOA estimation. Aiming to DOA estimation of coherent sources using the spatial smoothing technique and MUSIC algorithm, error analysis is first provided for the estimate of the corresponding mutual coupling coefficient matrix in a uniform linear array setting. An optimization formulation based on MUSIC is then established for joint estimation of mutual coupling coefficients and DOAs. The proposed IRA framework, which solves this optimization problem, has the advantages of few parameters and low computational complexity. Four different importance resampling algorithms which are the position resampling algorithm, the number resampling algorithm, the number and position resampling algorithm and the zero forcing resampling algorithm are proposed for the implement of the IRA framework with different arrangement and combination according to the characteristic of the real coupling coefficients vector. The proposed approach is effective not only to the case of coherent sources, but also for independent sources, and it can work in both single- and multiple-snapshot scenarios. The excellent performance of our methods for mutual coupling correction and DOA estimation is demonstrated via extensive simulation results, which can also be applied to other sensor parameter estimation scenarios.

Exploring Characteristics of Homicide Offenders With Schizophrenia Spectrum Disorders Via Machine Learning.

The link between schizophrenia and homicide has long been the subject of research with significant impact on mental health policy, clinical practice, and public perception of people with psychiatric disorders. The present study investigates factors contributing to completed homicides committed by offenders diagnosed with schizophrenia referred to a Swiss forensic institution, using machine learning algorithms. Data were collected from 370 inpatients at the Centre for Inpatient Forensic Therapy at the Zurich University Hospital of Psychiatry. A total of 519 variables were explored to differentiate homicidal and other (violent and non-violent) offenders. The dataset was split employing variable filtering, model building, and selection embedded in a nested resampling approach. Ten factors regarding criminal and psychiatric history and clinical factors were identified to be influential in differentiating between homicidal and other offenders. Findings expand the research on influential factors for completed homicide in patients with schizophrenia. Limitations, clinical relevance, and future directions are discussed.