Search

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]

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]

An issue for class‐imbalanced learning is what assessment metric should be employed. So far, precision‐recall curve (PRC) as a metric is rarely used in practice as compared with its alternative of receiver operating characteristic (ROC). This study investigates the performance of PRC as the evaluating criterion to address the class‐imbalanced data and focuses on the comparison of PRC with ROC. The advantages of PRC over ROC on assessing class‐imbalanced data are also investigated and tested on our proposed algorithm by tuning the whole model parameters in simulation studies and real data examples. The result shows that PRC is competitive with ROC as performance measurement for handling class‐imbalanced data in tuning the model parameters. PRC can be considered as an alternative but effective assessment for preprocessing (such as variable selection) skewed data and building a classifier in class‐imbalanced learning. [ABSTRACT FROM AUTHOR]