Artificial intelligence technology is rapidly developing with the improvement of computer performance and development various algorithms, research using artificial being actively conducted in field manufacturing technology. In welding, on arc welding quality prediction neural network algorithms (ANN) was mainly early stages. Since then, case a deep (DNN) algorithm, has been to increase accuracy by increasing hidden layer ANN algorithm. Recently, many studies have form predicting classifying...

Abstract We consider interval‐valued data that frequently appear with advanced technologies in current collection processes. Interval‐valued refer to the are observed as ranges instead of single values. In last decade, several approaches regression analysis have been introduced, but little work has done on relevant statistical inferences concerning model. this paper, we propose a new approach fit linear model using resampling idea. A key advantage is it enables one make such overall...

Summary We present a method for individual and integrative analysis of high dimension, low sample size data that capitalizes on the recurring theme in multivariate projecting higher dimensional onto few meaningful directions are solutions to generalized eigenvalue problem. propose general framework, called SELP (Sparse Estimation with Linear Programming), which one can obtain sparse estimate solution vector demonstrate utility canonical correlation an methylation gene expression profiles...

Journal Article Discriminant analysis through a semiparametric model Get access Y. Lin, Lin Search for other works by this author on: Oxford Academic Google Scholar Jeon Biometrika, Volume 90, Issue 2, June 2003, Pages 379–392, https://doi.org/10.1093/biomet/90.2.379 Published: 01 2003

Penalized likelihood density estimation provides an effective approach to the nonparametric fitting of graphical models, with conditional independence struc- tures characterized via selective term elimination in functional ANOVA decomposi- tions log density. A bottleneck has been cost numerical integration, which limited its application low-dimensional problems. In Jeon and Lin (2006), a reformulation was proposed replace multi-dimensional inte- grals by sums products univariate integrals,...

Summary Functional linear discriminant analysis provides a simple yet efficient method for classification, with the possibility of achieving perfect classification. Several methods have been proposed in literature that mostly address dimensionality problem. On other hand, there is growing interest interpretability analysis, which favours and sparse solution. In this paper we propose new approach incorporates type sparsity identifies nonzero subdomains functional setting, yielding solution...

This article concerns datasets in which variables are the form of intervals, obtained by aggregating information about from a larger dataset. We propose to view observed set hyper-rectangles as an empirical histogram, and use Gaussian kernel type estimator approximate its underlying distribution nonparametric way. apply this idea both univariate density estimation regression problems. Unlike many existing methods used analysis, proposed method can estimate conditional response variable for...

To assess the incremental prognostic value of coronary computed tomography angiography (CCTA) in comparison to a clinical risk model (Framingham score, FRS) and artery calcium score (CACS) for future cardiac events ischemic stroke patients without chest pain. This retrospective study included 1418 with acute who had no previous disease underwent CCTA, including CACS. Stenosis degree plaque types (high-risk, non-calcified, mixed, or calcified plaques) were assessed as CCTA variables....

Abstract We propose new discrimination methods for classification of high dimension, low sample size (HDLSS) data that regularize the degree piling. The within-class scatter HDLSS data, when projected onto a low-dimensional discriminant subspace, can be selected to arbitrarily small. Using this fact, we develop two different ways tuning amount scatter, or equivalently, In first approach, consider linear path connecting maximal piling and least directions. also formulate problem finding...

We propose a new algorithm for sparse estimation of eigenvectors in generalized eigenvalue problems (GEP). The GEP arises number modern data-analytic situations and statistical methods, including principal component analysis (PCA), multiclass linear discriminant (LDA), canonical correlation (CCA), sufficient dimension reduction (SDR) invariant co-ordinate selection. to modify the standard orthogonal iteration with sparsity-inducing penalty eigenvectors. To achieve this goal, we generalize...

There are a huge number of features which said to improve Convolutional Neural Network (CNN) accuracy.Practical testing combinations such on large datasets, and theoretical justification the result, is required.Some operate certain models exclusively for problems exclusively, or only small-scale datasets; while some features, as batch-normalization residual-connections, applicable majority models, tasks, datasets.We assume that universal include Weighted-Residual-Connections (WRC),...

We propose an exploratory data analysis approach when are observed as intervals in a nonparametric regression setting. The interval-valued contain richer information than single-valued the sense that they provide both center and range of underlying structure. Conventionally, these two attributes have been studied separately traditional tools can be readily used for analysis. unified tool attempts to capture relationship between response covariate by simultaneously accounting variability...

The Gross Motor Function Classification System (GMFCS) is a widely used tool for assessing the mobility of people with Cerebral Palsy (CP). It classifies patients into different levels based on their gross motor function and its level typically determined through visual evaluation by trained expert. Although gait analysis commonly in CP research, functional aspects patterns has yet to be fully exploited. By utilizing predict GMFCS, we can gain more comprehensive understanding how affects...

In this article, we introduce a characterization of the log-density smoothing spline ANOVA model. We show that in model order r (consisting main effects and all interactions up to r), joint density function is uniquely determined by collection r-dimensional marginal densities. Furthermore, largest under which marginals. Our results are valid for with other general structures.

This paper proposes a calibrated concave convex procedure (calibrated CCCP) for high-dimensional graphical model selection. The CCCP approach the smoothly clipped absolute deviation (SCAD) penalty is known to be path-consistent with probability converging one in linear regression models. We implement method SCAD use quadratic objective function undirected Gaussian models and adopt sparse estimation. For tuning procedure, we propose columnwise on adjusted test data. In simulation study,...

In multi-class discrimination with high-dimensional data, identifying a lower-dimensional subspace maximum class separation is crucial. We propose new optimization criterion for finding such discriminant subspace, which the ratio of two traces: trace between-class scatter matrix and within-class matrix. Since this problem not well-defined we to regularize within maximize between trace. A careful investigation reveals that has an innate connection eigenvalue decomposition indefinite For sake...

Functional linear discriminant analysis offers a simple yet efficient method for classification, with the possibility of achieving perfect classification. Several methods are proposed in literature that mostly address dimensionality problem. On other hand, there is growing interest interpretability analysis, which favors and sparse solution. In this work, we propose new approach incorporates type sparsity identifies non-zero sub-domains functional setting, offering solution easier to...

시뮬레이션은 현대의 복잡한 환경에서 효율적인 의사 결정을 위한 도구로서 폭넓게 사용되고 있다. 하지만, 대부분은 결과의 분석 보다는 시뮬레이션 모델의 개발 및 수행에 중점을 두고 본 논문에서는 모델링은 물론 분석도 중요하고 체계적으로 진행되어야 한다는 점을 강조하고, 이를 위하여 통계분석과 다양한 데이터 조작 그래픽 기능을 가진 R을 사용하여 이윤 최적화 시뮬레이션에 대한 모델링과 결과 데이터의 그래픽을 사용한 작업을 수행하여 유용성을 입증하였다. Simulation is now using in various area as an effective decision analysis tool complex environment of today. But, There a focus to the simulation model development and execution better than result analysis. This article will emphasis...