Jangsun Baek

Clustering in high-dimensional data becomes particularly challenging when noise deviates from symmetric, zero-mean distributions such as the Gaussian or Laplace distribution. In these cases, asymmetric non-Gaussian distributions provide a more accurate representation of the data's structure, especially when learning low-dimensional subspaces. However, selecting an appropriate family of asymmetric distributions is difficult when variables within each group exhibit varying degrees of skewness-some being right-skewed and others left-skewed. In this study, we propose a novel deep generative clustering model, ManlyMM-VAE, based on the variational autoencoder (VAE) framework. The model incorporates the Manly transformation, a parametric technique that flexibly adjusts the skewness of each variable, effectively handling both right- and left-skewed data. A Manly transformation mixture model serves as the prior in the latent space. The evidence lower bound (ELBO) is derived using the stochastic gradient variational Bayes (SGVB) estimator to optimize model parameters. We also introduce a cross-validation procedure for model selection. Experimental results on both synthetic and real-world datasets show that ManlyMM-VAE effectively clusters data with varying skewness, achieving significantly higher accuracy than competing methods. Furthermore, its robustness is confirmed through clustering experiments on image data corrupted to various degrees.

One of the challenges in clustering categorical data is the curse of dimensionality caused by the inherent sparsity of high-dimensional data, the records of which include a large number of attributes. The latent class model (LCM) assumes local independence between the variables in clusters, and is a parsimonious model-based clustering approach that has been used to circumvent the problem. The mixture of a log-linear model is more flexible but requires more parameters to be estimated. In this research, we recognize that each categorical observation can be conceived as a network with pairwise linked nodes, which are the response levels of the observation attributes. Therefore, the categorical data for clustering is considered a finite mixture of different component layer networks with distinct patterns. We apply a penalized composite likelihood approach to a finite mixture of networks for sparse multivariate categorical data to reduce the number of parameters, implement the EM algorithm to estimate the model parameters, and show that the estimates are consistent and satisfy asymptotic normality. The performance of the proposed approach is shown to be better in comparison with the conventional methods for both synthetic and real data sets.

Along with advances in technology, matrix data, such as medical/industrial images, have emerged in many practical fields. These data usually have high dimensions and are not easy to cluster due to their intrinsic correlated structure among rows and columns. Most approaches convert matrix data to multi dimensional vectors and apply conventional clustering methods to them, and thus, suffer from an extreme high-dimensionality problem as well as a lack of interpretability of the correlated structure among row/column variables. Recently, a regularized model was proposed for clustering matrix-valued data by imposing a sparsity structure for the mean signal of each cluster. We extend their approach by regularizing further on the covariance to cope better with the curse of dimensionality for large size images. A penalized matrix normal mixture model with lasso-type penalty terms in both mean and covariance matrices is proposed, and then an expectation maximization algorithm is developed to estimate the parameters. The proposed method has the competence of both parsimonious modeling and reflecting the proper conditional correlation structure. The estimators are consistent, and their limiting distributions are derived. We applied the proposed method to simulated data as well as real datasets and measured its clustering performance with the clustering accuracy (ACC) and the adjusted rand index (ARI). The experiment results show that the proposed method performed better with higher ACC and ARI than those of conventional methods.

Onion yield estimation using spatial panel regression model panel analysis;spatial econometric model;meteorological data;onion yields; Onions are grown in a few specific regions of Korea that depend on the climate and the regional characteristic of the production area. Therefore, when onion yields are to be estimated, it is reasonable to use a statistical model in which both the climate and the region are considered simultaneously. In this paper, using a spatial panel regression model, we predicted onion yields with the different weather conditions of the regions. We used the spatial auto regressive (SAR) model that reflects the spatial lag, and panel data of several climate variables for 13 main onion production areas from 2006 to 2015. The spatial weight matrix was considered for the model by the threshold value method and the nearest neighbor method, respectively. Autocorrelation was detected to be significant for the best fitted model using the nearest neighbor method. The random effects model was chosen by the Hausman test, and the significant climate variables of the model were the cumulative duration time of sunshine (January), the average relative humidity (April), the average minimum temperature (June), and the cumulative precipitation (November).

Motivation: Mixtures of factor analyzers enable model-based clustering to be undertaken for high-dimensional microarray data, where the number of observations n is small relative to the number of genes p. Moreover, when the number of clusters is not small, for example, where there are several different types of cancer, there may be the need to reduce further the number of parameters in the specification of the component-covariance matrices. A further reduction can be achieved by using mixtures of factor analyzers with common component-factor loadings (MCFA), which is a more parsimonious model. However, this approach is sensitive to both non-normality and outliers, which are commonly observed in microarray experiments. This sensitivity of the MCFA approach is due to its being based on a mixture model in which the multivariate normal family of distributions is assumed for the component-error and factor distributions. Results: An extension to mixtures of t-factor analyzers with common component-factor loadings is considered, whereby the multivariate t-family is adopted for the component-error and factor distributions. An EM algorithm is developed for the fitting of mixtures of common t-factor analyzers. The model can handle data with tails longer than that of the normal distribution, is robust against outliers and allows the data to be displayed in low-dimensional plots. It is applied here to both synthetic data and some microarray gene expression data for clustering and shows its better performance over several existing methods. Availability: The algorithms were implemented in Matlab. The Matlab code is available at http://blog.naver.com/aggie100. Contact:jbaek@jnu.ac.kr Supplementary information: Supplementary data are available at Bioinformatics online.

Mixtures of factor analyzers enable model-based density estimation to be undertaken for high-dimensional data, where the number of observations n is not very large relative to their dimension p. In practice, there is often the need to further reduce the number of parameters in the specification of the component-covariance matrices. To this end, we propose the use of common component-factor loadings, which considerably reduces further the number of parameters. Moreover, it allows the data to be displayed in low--dimensional plots.

Motivation: We present a new approach to the analysis of images for complementary DNA microarray experiments. The image segmentation and intensity estimation are performed simultaneously by adopting a two-component mixture model. One component of this mixture corresponds to the distribution of the background intensity, while the other corresponds to the distribution of the foreground intensity. The intensity measurement is a bivariate vector consisting of red and green intensities. The background intensity component is modeled by the bivariate gamma distribution, whose marginal densities for the red and green intensities are independent three-parameter gamma distributions with different parameters. The foreground intensity component is taken to be the bivariate t distribution, with the constraint that the mean of the foreground is greater than that of the background for each of the two colors. The degrees of freedom of this t distribution are inferred from the data but they could be specified in advance to reduce the computation time. Also, the covariance matrix is not restricted to being diagonal and so it allows for nonzero correlation between R and G foreground intensities. This gamma-t mixture model is fitted by maximum likelihood via the EM algorithm. A final step is executed whereby nonparametric (kernel) smoothing is undertaken of the posterior probabilities of component membership. The main advantages of this approach are: (1) it enjoys the well-known strengths of a mixture model, namely flexibility and adaptability to the data; (2) it considers the segmentation and intensity simultaneously and not separately as in commonly used existing software, and it also works with the red and green intensities in a bivariate framework as opposed to their separate estimation via univariate methods; (3) the use of the three-parameter gamma distribution for the background red and green intensities provides a much better fit than the normal (log normal) or t distributions; (4) the use of the bivariate t distribution for the foreground intensity provides a model that is less sensitive to extreme observations; (5) as a consequence of the aforementioned properties, it allows segmentation to be undertaken for a wide range of spot shapes, including doughnut, sickle shape and artifacts. Results: We apply our method for gridding, segmentation and estimation to cDNA microarray real images and artificial data. Our method provides better segmentation results in spot shapes as well as intensity estimation than Spot and spotSegmentation R language softwares. It detected blank spots as well as bright artifact for the real data, and estimated spot intensities with high-accuracy for the synthetic data. Availability: The algorithms were implemented in Matlab. The Matlab codes implementing both the gridding and segmentation/estimation are available upon request. Contact:jbaek@chonnam.ac.kr Supplementary information: Supplementary material is available at Bioinformatics online.