Search

Many real‐world data mining applications involve using imbalanced datasets to obtain predictive models. Imbalanced data can hinder the model performance of learning algorithms in rare cases. Although there are many well‐researched classification task solutions, most of them cannot be directly applied to regression task. One of the challenges in imbalanced regression is to find a suitable evaluation and optimization standard that can improve the predictive ability of the model without severe model bias. Based on the importance of rare cases, this study proposes a new evaluation metric called adapted squared error relevance (ASER) by defining new relevance function and weighting functions. This metric weights data points by defining the importance of rare cases and assigns different weights to losses of the same size at different rare cases, thus enabling the model selected by this evaluation metric to better predict rare cases. ASER is compared with SER on 32 real datasets and 9 simulated datasets to verify the predictive performance of the selected model at rare cases. The experimental results show that the new evaluation metric ASER can obtain a high prediction performance at rare cases, while also not losing too much prediction accuracy in common cases. In order to better evaluate the predictive ability of the model in rare cases, we propose a new evaluation metric for imbalanced regression called adapted squared error relevance (ASER). ASER weights data points by defining the importance of cases and assigns different weights to losses of the same size at different cases. ASER can obtain a high prediction performance at rare cases, while also not losing too much prediction accuracy in common cases. [ABSTRACT FROM AUTHOR]

When developing prediction models for small or sparse binary data with many highly correlated covariates, logistic regression often encounters separation or multicollinearity problems, resulting serious bias and even the nonexistence of standard maximum likelihood estimates. The combination of separation and multicollinearity makes the task of logistic regression more difficult, and a few studies addressed separation and multicollinearity simultaneously. In this paper, we propose a double-penalized method called lFRE to combat separation and multicollinearity in logistic regression. lFRE combines the logF-type penalty with the ridge penalty. The results indicate that compared with other penalty methods, lFRE can not only effectively remove bias from predicted probabilities but also provide the minimum mean squared prediction error. Aside from that, a real dataset is also employed to test the performance of the lFRE algorithm compared with several existing methods. The result shows that lFRE has strong competitiveness compared with them and can be used as an alternative algorithm in logistic regression to solve separation and multicollinearity problems. [ABSTRACT FROM AUTHOR]

Aims: Attention‐deficit/hyperactivity disorder (ADHD) is a neuropsychiatric disorder of substantial heritability, yet emerging evidence suggests that key risk variants might reside in the noncoding regions of the genome. Our study explored the association of lncRNAs (long noncoding RNAs) with ADHD as represented at three different phenotypic levels guided by the Research Domain Criteria (RDoC) framework: (i) ADHD caseness and symptom dimension, (ii) executive functions as functional endophenotype, and (iii) potential genetic influence on white matter architecture as brain structural endophenotype. Methods: Genotype data of 107 tag single nucleotide polymorphisms (SNP) from 10 candidate lncRNAs were analyzed in 1040 children with ADHD and 630 controls of Chinese Han descent. Executive functions including inhibition and set‐shifting were assessed by STROOP and trail making tests, respectively. Imaging genetic analyses were performed in a subgroup of 33 children with ADHD and 55 controls using fractional anisotropy (FA). Results: One SNP rs3908461 polymorphism in RNF219‐AS1 was found to be significantly associated with ADHD caseness: with C‐allele detected as the risk genotype in the allelic model (P = 8.607E‐05) and dominant genotypic model (P = 9.628E‐05). Nominal genotypic effects on inhibition (p = 0.020) and set‐shifting (p = 0.046) were detected. While no direct effect on ADHD core symptoms was detected, mediation analysis suggested that SNP rs3908461 potentially exerted an indirect effect through inhibition function [B = 0.21 (SE = 0.12), 95% CI = 0.02‐0.49]. Imaging genetic analyses detected significant associations between rs3908461 genotypes and FA values in corpus callosum, left superior longitudinal fasciculus, left posterior limb of internal capsule, left posterior thalamic radiate (include optic radiation), and the left anterior corona radiate (PFWE corrected < 0.05). Conclusion: Our present study examined the potential roles of lncRNA in genetic etiological of ADHD and provided preliminary evidence in support of the potential RNF219‐AS1 involvement in the pathophysiology of ADHD in line with the RDoC framework. [ABSTRACT FROM AUTHOR]

Background: Feature selection in class-imbalance learning has gained increasing attention in recent years due to the massive growth of high-dimensional class-imbalanced data across many scientific fields. In addition to reducing model complexity and discovering key biomarkers, feature selection is also an effective method of combating overlapping which may arise in such data and become a crucial aspect for determining classification performance. However, ordinary feature selection techniques for classification can not be simply used for addressing class-imbalanced data without any adjustment. Thus, more efficient feature selection technique must be developed for complicated class-imbalanced data, especially in the context of high-dimensionality. Results: We proposed an algorithm called sssHD to achieve stable sparse feature selection applied it to complicated class-imbalanced data. sssHD is based on the Hellinger distance (HD) coupled with sparse regularization techniques. We stated that Hellinger distance is not only class-insensitive but also translation-invariant. Simulation result indicates that HD-based selection algorithm is effective in recognizing key features and control false discoveries for class-imbalance learning. Five gene expression datasets are also employed to test the performance of the sssHD algorithm, and a comparison with several existing selection procedures is performed. The result shows that sssHD is highly competitive in terms of five assessment metrics. In addition, sssHD presents limited differences between performing and not performing re-balance preprocessing. Conclusions: sssHD is a practical feature selection method for high-dimensional class-imbalanced data, which is simple and can be an alternative for performing feature selection in class-imbalanced data. sssHD can be easily extended by connecting it with different re-balance preprocessing, different sparse regularization structures as well as different classifiers. As such, the algorithm is extremely general and has a wide range of applicability. [ABSTRACT FROM AUTHOR]

Elastic net (Enet) and sparse partial least squares (SPLS) are frequently employed for wavelength selection and model calibration in analysis of near infrared spectroscopy data. Enet and SPLS can perform variable selection and model calibration simultaneously. And they also tend to select wavelength intervals rather than individual wavelengths when the predictors are multicollinear. In this paper, we focus on comparison of Enet and SPLS in interval wavelength selection and model calibration for near infrared spectroscopy data. The results from both simulation and real spectroscopy data show that Enet method tends to select less predictors as key variables than SPLS; thus it gets more parsimony model and brings advantages for model interpretation. SPLS can obtain much lower mean square of prediction error (MSE) than Enet. So SPLS is more suitable when the attention is to get better model fitting accuracy. The above conclusion is still held when coming to performing the strongly correlated NIR spectroscopy data whose predictors present group structures, Enet exhibits more sparse property than SPLS, and the selected predictors (wavelengths) are segmentally successive. [ABSTRACT FROM AUTHOR]

Feature selection or elimination methods have been widely used to mine high-dimensional data from many fields such as biomedicine. However, classical feature selection methods usually focus on class-balanced and numerical data, and the classification performance is poor when processing class-imbalanced or mixed-type data. Moreover, existing studies hardly consider the co-occurrence of both class-imbalance and mixture of numerical and nominal features. In this paper, we propose an adaptive loss backward feature elimination method for class-imbalanced and mixed-type data (FEIM). In FEIM, a technique to convert and decompose nominal features is employed to process mixed-type data. Meanwhile, an adaptive loss function is proposed to make the classifier pay more attention to the minority class. In experiments, the effectiveness and stability of FEIM are synthetically evaluated against seven medical diagnostic datasets in terms of the measurements of F measurement and precision, and the results of feature selection are analyzed in detail. The experimental results show that, compared with three common methods (Recursive Feature Elimination SVM (RFE-SVM), Relief, and Holdout Algorithm for Backward Feature Elimination (HO-BFE)), FEIM can achieve a higher classification accuracy and is more stable when the parameters are changed. On the other hand, the feature selection results of FEIM are more meaningful by processing class-imbalanced and mixed-type data. Therefore, FEIM is a competitive and effective method to process feature elimination for class-imbalanced and mixed-type biomedical data for in medical diagnosis. The supporting information is available in the electronic supplementary material. • A mixed-type feature selection algorithms called FEIM is proposed for imbalanced data with both numerical and nominal features. • An adaptive loss function is designed for class-imbalance learning. • FEIM is more stable when the model parameters are changed compared with several classical methods. [ABSTRACT FROM AUTHOR]

Screening important variables for class-imbalanced data is still a challenging task. In this study, we propose an algorithm for stably selecting key variables on class-imbalanced data based on the precision-recall curve (PRC), where the PRC is utilized as the assessment criterion in the model building stage, and sparse regularized logistic regression combined with subsampling (SRLRS) is designed to perform stable variable selection. Considering the characteristic of class-imbalanced data, we also proposed classification-based partition for cross validation, as well as leaving half of majority observations out and leaving one minority observation out (LHO-LOO) for subsampling. Simulation results and real data showed that our algorithm is highly suitable for handling class-imbalanced data, and that the PRC can be an alternative evaluation criterion for model selection when handling class-imbalanced data. [ABSTRACT FROM AUTHOR]

The partial least squares-discriminant analysis (PLS-DA) is the most widely used statistical tool to perform classification and biomarker screening in metabolomics. However, the PLS-DA tends to overfit the data, and the selection of biomarkers is often unstable because of the disturbance of uninformative variables in principal components. In this paper, we propose an algorithm for performing stable biomarker screening and for seeking the optimal generalization performance, in which the biomarker identification is based on sparse regularization variable selection in combination with subsampling (SRS), and the classification is subsequently performed by a linear support vector machine (SVM) classifier in the selected-variable space to obtain the maximum classification accuracy. Two metabolomics datasets measured by gas chromatography-mass spectrometry are employed to evaluate the performance of the proposed SRS-SVM algorithm, and the comparison with existing related algorithms is given. The result shows that the SRS-SVM algorithm outperforms the PLS-DA and is competitive with other related algorithms in terms of prediction classification accuracy measured by both internal and external validation. Furthermore, the selection of candidate biomarkers is quite stable by the SRS-SVM algorithm, and it can be an alternative and competitive method for the analysis of metabolomics data. The R code for implementing the SRS-SVM algorithm is available in the Electronic supplementary material . [ABSTRACT FROM AUTHOR]

This paper is the generalization of weight-fused elastic net (Fu and Xu, 2012), which performs group variable selection by combining weight-fused LASSO(wfLasso) and elastic net (Zou and Hastie, 2005) penalties. In this study, the elastic net penalty is replaced by adaptive elastic net penalty (AdaEnet) (Zou and Zhang, 2009), and a new group variable selection algorithm with oracle property (Fan and Li, 2001; Zou, 2006) is obtained. [ABSTRACT FROM PUBLISHER]

In this article, we consider the problem of variable selection and estimation with the strongly correlated multi-collinear data by using grouping variable selection techniques. A new grouping variable selection method, called weight-fused elastic net(WFEN), is proposed to deal with the high dimensional collinear data. The proposed model, combined two different grouping effect mechanisms induced by the elastic net and weight-fused LASSO, respectively, can be easily unified in the frame of LASSO and computed efficiently. The performance with the simulation and real data sets shows that our method is competitive with other related methods, especially when the data present high multi-collinearity. [ABSTRACT FROM PUBLISHER]

In this paper a novel wavelength region selection algorithm, called elastic net grouping variable selection combined with partial least squares regression (EN-PLSR), is proposed for multi-component spectral data analysis. The EN-PLSR algorithm can automatically select successive strongly correlated prediction variable groups related to the response variable using two steps. First, a portion of the correlated predictors are selected and divided into subgroups by means of the grouping effect of elastic net estimation. Then, a recursive leave-one-group-out strategy is employed to further shrink the variable groups in terms of the root mean square error of cross-validation (RMSECV) criterion. The performance of the algorithm with real near-infrared (NIR) spectroscopic data sets shows that the EN-PLSR algorithm is competitive with full-spectrum PLS and moving window partial least squares (MWPLS) regression methods and it is suitable for use with strongly correlated spectroscopic data. [ABSTRACT FROM AUTHOR]

Learning from class-imbalanced data has gained increasing attention in recent years due to the massive growth of skewed data across many scientific fields such as metabolomics. Some researches show that it is not the imbalance itself which hinders the classification performance, but class overlapping do play an important role in the performance degradation when associated to class-imbalance. So alleviating the overlapping of the imbalanced data might be an effective way to improve the performance in class-imbalance learning. In this study, we propose two feature selection algorithms that aim to minimize the overlap degree between the majority and the minority, which is based on a simple assumption that decreasing overlap degree of a data set makes it more separable. The proposed MOSNS and MOSS methods are built via sparse regularization techniques. Simulation results indicate that our algorithms is effective in recognizing key features and control false discoveries for class-imbalance learning. Four class-imbalanced metabolomics data sets are also employed to test the performance of our algorithm, and a comparison with accuracy (ACC)-based and ROC-based selection procedures is performed. The result shows that our algorithms are highly competitive and can be an alternative feature selection strategy in class-imbalance learning. ● Two feature selection algorithms are proposed by minimizing the overlap of a data set. ● Sparse feature selection strategy are used in class-imbalance learning. ● The classification performance of metabolomics data can be improved by sparse feature selection and re-balance preprocessing. [ABSTRACT FROM AUTHOR]

Feature selection and rebalancing can be seen as two preprocessing ways in class‐imbalanced learning. Recently, there have been many research achievements and applications on LASSO‐type feature selection, whereas most of them are not directly designed for addressing class‐imbalanced data. In this study, we proposed a LASSO‐based stable feature selection algorithm for class‐imbalanced data analysis, and false‐positive selection (FPS) under balanced and imbalanced situations was calculated via selection frequency of each predictor in doing stable selection. The results on simulation studies and real data examples show that class imbalance contributes to avoid overselection caused by LASSO when the data are highly correlated and a lower FPS can be obtained with class‐imbalanced data than balanced one in most of cases in the same settings. A statistical explanation was given for this phenomenon. In addition, it does not need to rebalance the class‐imbalanced data for performing such LASSO‐based feature selection with a stable strategy, and to some degree, intentionally disequilibrating the balanced data could be an alternative strategy to weaken overselection and to perform biomarker identification for finding a few of most important biomarkers. A LASSO‐based stable feature selection algorithm was proposed for imbalanced data. It is not necessary to rebalance the imbalanced data with such stable feature selection strategy. Class imbalance even contributes to avoid overselection caused by LASSO when data are strongly correlated. [ABSTRACT FROM AUTHOR]

Imbalanced domain prediction analysis is currently one of the hot research topics. Many real‐world data mining analyses involve using imbalanced data to obtain predictive models. In the context of imbalance, research on classification problems has been extensive, but research on regression problems is negligible. Rare values rarely occur in imbalanced regression problems, but the focus is on accurately predicting the continuous target variables of rare instances. One of the challenges in imbalanced regression is finding a suitable strategy to rebalance the original dataset in order to improve the predictive performance of the model in rare instances. In this study, two algorithms are proposed: sigma nearest over‐sampling based on convex combination for regression (SNOCCR) and affine combination‐based over‐sampling (ACOS). ACOS rebalances the original dataset by generating new instances through the affine combinations of the original examples. The region where the new instances are generated can be adjusted based on the distribution of the data, ensuring that the generated cases better mimic the distribution of the original examples. The comparison among ACOS, SNOCCR, and other preprocessing methods was conducted on 15 datasets to validate the predictive performance of models trained on rebalanced datasets for rare instances. The experimental results indicate that ACOS outperforms other existing methods. In this study, two over‐sampling algorithms are proposed: sigma nearest over‐sampling based on convex combination for regression (SNOCCR) and affine combination‐based over‐sampling (ACOS) in order to improve the predictive performance of the model in imbalanced regression. The experimental results indicate that ACOS outperforms other existing methods. [ABSTRACT FROM AUTHOR]

In class-imbalance learning, Synthetic Minority Oversampling Technique (SMOTE) is a widely used technique to tackle class-imbalance problems from the data level, whereas SMOTE blindly selects neighboring minority class points when performing an interpolation among them and inevitably brings collinearity between the generated new points and the original ones. To combat these problems, we propose in this study an adaptive-weighting SMOTE method, termed as AWSMOTE. AWSMOTE applies two types of SVM-based weights into SMOTE. A kind of weight is used in variable space to combat the drawbacks of collinearity, while another weight is utilized in sample space to purposefully choose those support vectors from the minority class as the neighboring points in the interpolation. AWSMOTE is compared with SMOTE and its improved versions with six simulated datasets and 22 real-world datasets. The results demonstrate the effectiveness and advantages of the proposed approach. [ABSTRACT FROM AUTHOR]

In class-imbalance learning, Synthetic Minority Oversampling Technique (SMOTE) is a widely used technique to tackle class-imbalance problems from the data level, whereas SMOTE blindly selects neighboring minority class points when performing an interpolation among them and inevitably brings collinearity between the generated new points and the original ones. To combat these problems, we propose in this study an adaptive-weighting SMOTE method, termed as AWSMOTE. AWSMOTE applies two types of SVM-based weights into SMOTE. A kind of weight is used in variable space to combat the drawbacks of collinearity, while another weight is utilized in sample space to purposefully choose those support vectors from the minority class as the neighboring points in the interpolation. AWSMOTE is compared with SMOTE and its improved versions with six simulated datasets and 22 real-world datasets. The results demonstrate the effectiveness and advantages of the proposed approach. [ABSTRACT FROM AUTHOR]

In class-imbalance learning, Synthetic Minority Oversampling Technique (SMOTE) is a widely used technique to tackle class-imbalance problems from the data level, whereas SMOTE blindly selects neighboring minority class points when performing an interpolation among them and inevitably brings collinearity between the generated new points and the original ones. To combat these problems, we propose in this study an adaptive-weighting SMOTE method, termed as AWSMOTE. AWSMOTE applies two types of SVM-based weights into SMOTE. A kind of weight is used in variable space to combat the drawbacks of collinearity, while another weight is utilized in sample space to purposefully choose those support vectors from the minority class as the neighboring points in the interpolation. AWSMOTE is compared with SMOTE and its improved versions with six simulated datasets and 22 real-world datasets. The results demonstrate the effectiveness and advantages of the proposed approach. [ABSTRACT FROM AUTHOR]

In class-imbalance learning, Synthetic Minority Oversampling Technique (SMOTE) is a widely used technique to tackle class-imbalance problems from the data level, whereas SMOTE blindly selects neighboring minority class points when performing an interpolation among them and inevitably brings collinearity between the generated new points and the original ones. To combat these problems, we propose in this study an adaptive-weighting SMOTE method, termed as AWSMOTE. AWSMOTE applies two types of SVM-based weights into SMOTE. A kind of weight is used in variable space to combat the drawbacks of collinearity, while another weight is utilized in sample space to purposefully choose those support vectors from the minority class as the neighboring points in the interpolation. AWSMOTE is compared with SMOTE and its improved versions with six simulated datasets and 22 real-world datasets. The results demonstrate the effectiveness and advantages of the proposed approach. [ABSTRACT FROM AUTHOR]

Innovations in genomics have enabled the development of low‐cost, high‐resolution, single nucleotide polymorphism (SNP) genotyping arrays that accelerate breeding progress and support basic research in crop science. Here, we developed and validated the SoySNP618K array (618,888 SNPs) for the important crop soybean. The SNPs were selected from whole‐genome resequencing data containing 2,214 diverse soybean accessions; 29.34% of the SNPs mapped to genic regions representing 86.85% of the 56,044 annotated high‐confidence genes. Identity‐by‐state analyses of 318 soybeans revealed 17 redundant accessions, highlighting the potential of the SoySNP618K array in supporting gene bank management. The patterns of population stratification and genomic regions enriched through domestication were highly consistent with previous findings based on resequencing data, suggesting that the ascertainment bias in the SoySNP618K array was largely compensated for. Genome‐wide association mapping in combination with reported quantitative trait loci enabled fine‐mapping of genes known to influence flowering time, E2 and GmPRR3b, and of a new candidate gene, GmVIP5. Moreover, genomic prediction of flowering and maturity time in 502 recombinant inbred lines was highly accurate (>0.65). Thus, the SoySNP618K array is a valuable genomic tool that can be used to address many questions in applied breeding, germplasm management, and basic crop research. [ABSTRACT FROM AUTHOR]

Objective Previous animal studies showed contradictory clinical observations on whether acute hyperglycemia contributes to poor outcome in traumatic brain injury (TBI). Herein, we tried to clarify this issue. Methods Striking with depths of 3.0–4.25 mm at right occipitoparietal brain region and with depth of 3.75 mm at right/left occipitoparietal or right/left frontoparietal brain region were performed, respectively. Blood glucose and insulin levels were traced every four hours from 1 to 72 h after striking. HOMA2-%S and HOMA2-%β were calculated. Modified neurological severity scores (mNSS) were used to evaluate neurological deficit within 72 h. Results Striking with depths of 3.5–4.25 mm induced increase in blood glucose lasting up to 24 h after striking. The levels of blood glucose after striking with depths of 3.75–4.25 mm were significantly different from that of striking with the depth of 3.0 mm. Striking with depth of 3.75 mm at right/left occipitoparietal region induced higher blood glucose in 24 h than that at right/left frontoparietal region. Insulin concentration increased slowly during 72 h after striking. Striking also induced decrease in insulin sensitivity and secretion lasting 72 h. Evaluation of mNSS revealed that severe striking (beyond 3.75 mm) worsened nerve function than slight striking (<3.0 mm). Intervention of acute hyperglycemia could decrease the mNSS from 2 to 7 days after TBI. Conclusion Our results suggested that only severe TBI could induce acute hyperglycemia by itself, and early care of acute hyperglycemia could benefit the outcome of TBI patients. [ABSTRACT FROM AUTHOR]

Key Clinical Message Residual periosteum developed periosteal bone formation in the pocket 10 years after cranioplasty, lumpectomy was conducted on the left lower abdomen under local anesthesia. Pathological sections revealed abundant osteocytes and mature bone matrix, and confirmed the bone formation on the residual periosteum. [ABSTRACT FROM AUTHOR]

Abstract: Kernel methods, in particular support vector machines, have been further extended into a new class of methods, which could effectively solve nonlinear problems in chemistry by using simple linear transformation. In fact, the kernel function used in kernel methods might be regarded as a general protocol to deal with nonlinear data in chemistry. In this paper, the basic idea and modularity of kernel methods, together with some simple examples, are discussed in detail to give an in-depth understanding for kernel methods. Three key ingredients of kernel methods, namely dual form, nonlinear mapping and kernel function, provide a consistent framework of kernel-based algorithms. The modularity of kernel methods allows linear algorithms to combine with any kernel function. Thus, some commonly used chemometric algorithms are easily extended to their kernel versions. [Copyright &y& Elsevier]