We present a novel approach for the analysis of transcriptomics data that integrates functional annotation of gene sets with expression values in a multivariate fashion, and directly assesses the relation of functional features to a multivariate space of response phenotypical variables. Multivariate projection methods are used to obtain new correlated variables for a set of genes that share a given function. These new functional variables are then related to the response variables of interest. The analysis of the principal directions of the multivariate regression allows for the identification of gene function features correlated with the phenotype. Two different transcriptomics studies are used to illustrate the statistical and interpretative aspects of the methodology. We demonstrate the superiority of the proposed method over equivalent approaches.

Modern industrial processes are characterized by acquiring massive amounts of highly collinear data. In this context, partial least-squares (PLS) regression, if wisely used, can become a strategic tool for process improvement and optimization. In this paper we illustrate the versatility of this technique through several real case studies that basically differ in the structure of the X matrix (process variables) and Y matrix (response parameters). By using the PLS approach, the results show that it is possible to build predictive models (soft sensors) for monitoring the performance of a wastewater treatment plant, to help in the diagnosis of a complex batch polymerization process, to develop an automatic classifier based on image data, or to assist in the empirical model building of a continuous polymerization process. Copyright © 2008 John Wiley & Sons, Ltd.

This paper presents a novel approach to the question of surface grading, the soft color texture descriptors method. This method is extracted from an extensive evaluation process of several factors based on the use of two well established statistical tools: experimental design and logistic regression. The utility of different combinations of factors is evaluated in regard to the problem of automatic classification of materials such as ceramic tiles that need to be grouped according to homogeneous visual appearance, that is, the surface grading application. The set of factors includes the number of neighbors in the k-NN classifier (several values of k parameter), color space representation schemes (CIE Lab, CIE Luv, RGB, and grayscale), and color texture features (mean, standard deviation, 2nd–5th histogram moments). A factorial experimental design is performed testing all combinations of the above factors on a large image database of ceramic tiles. Accuracy estimates are computed using logistic regression to determine the best combinations of factors. From the point of view of machine learning the overall process conforms a wrapper approach able to select significant design choices (k parameter in k-NN classifier and color space) and carry out a feature selection within the set of color texture features at the same time. Experiments were repeated with alternate color texture schemes from the literature: color histograms and centile-LBP. Comparisons of methods are presented describing both accuracy estimates and runtimes.

In this paper we present a new approach for the detection of defects in random colour textures. This approach is based on the use of the T2 statistic and it is derived from the MIA strategy (Multivariate Image Analysis) developed in recent years in the field of applied statistics. PCA analysis is used to extract a reference eigenspace from a matrix built by unfolding the RGB raw data of defect-free images. The unfolding is performed compiling colour and spatial information of pixels. New testing images are also unfolded and projected onto the reference eigenspace obtaining a score matrix used to compute the T2 images. These images are converted into defect maps which allow the location of defective pixels. Only very few samples are needed to perform unsupervised training. With regard to literature, the method uses one of the simplest approaches providing low computational costs.