• Communications for Statistical Applications and Methods
• /
• v.3 no.3
• /
• pp.25-30
• /
• 1996

In discriminant analysis the procedures commonly used to estimate the dimensionality involve testing a sequence of dimensionality hypotheses. There is a problem with the size of the test since dimensionality hypotheses are tested sequentially and thus they are actually conditional tests. The focus of this paper is to investigate in asymptotic sense what happens to the sequential testing procedure if the assumption of normality does not hold.

• International Journal of Computer Science & Network Security
• /
• v.21 no.3
• /
• pp.31-36
• /
• 2021

RNA sequencing (RNA-Seq) is a technology that facilitates transcriptome analysis using next-generation sequencing (NSG) tools. Information on the quantity and sequences of RNA is vital to relate our genomes to functional protein expression. RNA-Seq data are characterized as being high-dimensional in that the number of variables (i.e., transcripts) far exceeds the number of observations (e.g., experiments). Given the wide range of dimensionality reduction techniques, it is not clear which is best for RNA-Seq data analysis. In this paper, we study the effect of three dimensionality reduction techniques to improve the classification of the RNA-Seq dataset. In particular, we use PCA, SVD, and SOM to obtain a reduced feature space. We built nine classification models for a cancer dataset and compared their performance. Our experimental results indicate that better classification performance is obtained with PCA and SOM. Overall, the combinations PCA+KNN, SOM+RF, and SOM+KNN produce preferred results.

• The Pure and Applied Mathematics
• /
• v.3 no.2
• /
• pp.141-154
• /
• 1996

A data-driven index of dimensionality for an educational or psychological test - DETECT, short for Dimensionality Evaluation To Enumerate Contributing Traits, is proposed in this paper. It is based on estimated conditional covariances of item pairs, given score on remaining test items. Its purpose is to detect whatever multidimensionality structure exists, especially in the case of approximate simple structure. It does so by assigning items to relatively dimensionally homogeneous clusters via attempted maximization of the DETECT over all possible item cluster partitions. The performance of DETECT is studied through real and simulated data analyses.

• Journal of Information Processing Systems
• /
• v.12 no.3
• /
• pp.468-488
• /
• 2016

The paper proposes a novel framework for 3D face verification using dimensionality reduction based on highly distinctive local features in the presence of illumination and expression variations. The histograms of efficient local descriptors are used to represent distinctively the facial images. For this purpose, different local descriptors are evaluated, Local Binary Patterns (LBP), Three-Patch Local Binary Patterns (TPLBP), Four-Patch Local Binary Patterns (FPLBP), Binarized Statistical Image Features (BSIF) and Local Phase Quantization (LPQ). Furthermore, experiments on the combinations of the four local descriptors at feature level using simply histograms concatenation are provided. The performance of the proposed approach is evaluated with different dimensionality reduction algorithms: Principal Component Analysis (PCA), Orthogonal Locality Preserving Projection (OLPP) and the combined PCA+EFM (Enhanced Fisher linear discriminate Model). Finally, multi-class Support Vector Machine (SVM) is used as a classifier to carry out the verification between imposters and customers. The proposed method has been tested on CASIA-3D face database and the experimental results show that our method achieves a high verification performance.

• Korean Journal of Remote Sensing
• /
• v.35 no.6_1
• /
• pp.959-971
• /
• 2019

CNN (Convolutional Neural Network) is one representative deep learning algorithm, which can extract high-level spatial and spectral features, and has been applied for hyperspectral image classification. However, one significant drawback behind the application of CNNs in hyperspectral images is the high dimensionality of the data, which increases the training time and processing complexity. To address this problem, several CNN based hyperspectral image classification studies have exploited PCA (Principal Component Analysis) for dimensionality reduction. One limitation to this is that the spectral information of the original image can be lost through PCA. Although it is clear that the use of PCA affects the accuracy and the CNN training time, the impact of PCA for CNN based hyperspectral image classification has been understudied. The purpose of this study is to analyze the quantitative effect of PCA in CNN for hyperspectral image classification. The hyperspectral images were first transformed through PCA and applied into the CNN model by varying the size of the reduced dimensionality. In addition, 2D-CNN and 3D-CNN frameworks were applied to analyze the sensitivity of the PCA with respect to the convolution kernel in the model. Experimental results were evaluated based on classification accuracy, learning time, variance ratio, and training process. The size of the reduced dimensionality was the most efficient when the explained variance ratio recorded 99.7%~99.8%. Since the 3D kernel had higher classification accuracy in the original-CNN than the PCA-CNN in comparison to the 2D-CNN, the results revealed that the dimensionality reduction was relatively less effective in 3D kernel.

• Communications for Statistical Applications and Methods
• /
• v.13 no.3
• /
• pp.467-477
• /
• 2006

A multidimensionality detecting procedure DETECT, based on conditional covariances between items, is extended and refined to deal with polytomous item data as well as binary one. A large body of simulation study shows extraordinary performance of DETECT in both enumerating degrees of multidimensionality in a test and discovering dimensionally distinctive item clusters. Real data study also provides very meaningful results, making DETECT a strong dimensionality assessment tool for the test data analysis.

• Communications for Statistical Applications and Methods
• /
• v.15 no.2
• /
• pp.265-271
• /
• 2008

It is commonly believed that diseases of human or economic traits of livestock are caused not by single genes acting alone, but multiple genes interacting with one another. This issue is difficult due to the limitations of parametric-statistic methods of gene effects. So we introduce multifactor-dimensionality reduction(MDR) as a methods for reducing the dimensionality of multilocus information. The MDR method is nonparametric (i. e., no hypothesis about the value of a statistical parameter is made), model free (i. e., it assumes no particular inheritance model) and is directly applicable to case-control studies. Application of the MDR method revealed the best model with an interaction effect between the SNPs, SNP1 and SNP3, while only one main effect of SNP1 was statistically significant for LMA (p < 0.01) under a general linear mixed model.

• Genomics & Informatics
• /
• v.16 no.4
• /
• pp.37.1-37.3
• /
• 2018

Gene-gene interaction is a key factor for explaining missing heritability. Many methods have been proposed to identify gene-gene interactions. Multifactor dimensionality reduction (MDR) is a well-known method for the detection of gene-gene interactions by reduction from genotypes of single-nucleotide polymorphism combinations to a binary variable with a value of high risk or low risk. This method has been widely expanded to own a specific objective. Among those expansions, fuzzy-MDR uses the fuzzy set theory for the membership of high risk or low risk and increases the detection rates of gene-gene interactions. Fuzzy-MDR is expanded by a maximum likelihood estimator as a new membership function in empirical fuzzy MDR (EFMDR). However, EFMDR is relatively slow, because it is implemented by R script language. Therefore, in this study, we implemented EFMDR using RCPP ($c^{{+}{+}}$ package) for faster executions. Our implementation for faster EFMDR, called EMMDR-Fast, is about 800 times faster than EFMDR written by R script only.

• BMB Reports
• /
• v.42 no.9
• /
• pp.617-622
• /
• 2009

We investigated the genetic associations of ischemic stroke by identifying epistasis of its heterogeneous subtypes such as small vessel occlusion (SVO) and large artery atherosclerosis (LAA). Epistasis was analyzed with 24 genes in 207 controls and 271 patients (SVO = 110, LAA = 95) using multifactor dimensionality reduction and entropy decomposition. The multifactor dimensionality reduction analysis with any of 1- to 4-locus models showed no significant association with LAA (P > 0.05). The analysis of SVO, however, revealed a significant association in the best 3-locus model with P10L of TGF-$\beta{1}$, C1013T of SPP1, and R485K of F5 (testing balanced accuracy = 63.17%, P < 0.05). Subsequent entropy analysis also revealed that such heterogeneity was present and quite a large entropy was estimated among the 3 loci for SVO (5.43%), but only a relatively small entropy was estimated for LAA (1.81%). This suggests that the synergistic epistasis model might contribute specifically to the pathogenetsis of SVO, which implies a different etiopathogenesis of the ischemic stroke subtypes.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.25 no.3
• /
• pp.219-230
• /
• 2022

In implementing a robust automatic target recognition(ATR) system with synthetic aperture radar(SAR) imagery, one of the most important issues is accurate classification of target variants, which are the same targets with different serial numbers, configurations and versions, etc. In this paper, a deep learning network with channel attention modules is proposed to cope with the recognition problem for target variants based on the previous research findings that the channel attention mechanism selectively emphasizes the useful features for target recognition. Different from other existing attention methods, this paper employs the channel attention modules without dimensionality reduction along the channel direction from which direct correspondence between feature map channels can be preserved and the features valuable for recognizing SAR target variants can be effectively derived. Experiments with the public benchmark dataset demonstrate that the proposed scheme is superior to the network with other existing channel attention modules.