SELECTION OF FREE PARAMETER FOREL-2 ALGORITHM IN THE PROBLEM OF AUTOMATIC GROUPING OF INDUSTRIAL PRODUCTS BY HOMOGENEOUS PRODUCTION BATCHES

Исследуется проблема влияния полуавтоматического подбора свободного параметра в задаче автоматической группировки промышленной продукции по однородным производственным партиям полупроводниковых приборов, основанной на модели FOREL-2 для задач p-медиан и k-средних. The author considers the problem impact of semi-automatic selection free parameter in the problem of automatic grouping of industrial products by homogeneous production batches based on the FOREL-2 model. We provide a comparative automatic grouping quality assessment results with models of FOREL-2, p-median and k-means.

COMPARATIVE ANALYSIS OF THE K-STANDARDS ALGORITHM APPLICATION FOR THE CLUSTERING PROBLEM

В статье рассматривается применение алгоритма k-эталонов для задачи кластеризации на примере производственных партий электрорадиоизделий, сделан вывод о качестве работы алгоритма k-эталонов и целесообразности его использования при решении задач автоматической группировки продукции. The article discusses the application of the k-standards algorithm for the clustering problem on the example of production batches of electrical radio products, a conclusion was made about the quality of the k-standards algorithm and the expediency of its use in automatic grouping problems solving.

Pairwise Elastic Net Representation-Based Classification for Hyperspectral Image Classification

The representation-based algorithm has raised a great interest in hyperspectral image (HSI) classification. l1-minimization-based sparse representation (SR) attempts to select a few atoms and cannot fully reflect within-class information, while l2-minimization-based collaborative representation (CR) tries to use all of the atoms leading to mixed-class information. Considering the above problems, we propose the pairwise elastic net representation-based classification (PENRC) method. PENRC combines the l1-norm and l2-norm penalties and introduces a new penalty term, including a similar matrix between dictionary atoms. This similar matrix enables the automatic grouping selection of highly correlated data to estimate more robust weight coefficients for better classification performance. To reduce computation cost and further improve classification accuracy, we use part of the atoms as a local adaptive dictionary rather than the entire training atoms. Furthermore, we consider the neighbor information of each pixel and propose a joint pairwise elastic net representation-based classification (J-PENRC) method. Experimental results on chosen hyperspectral data sets confirm that our proposed algorithms outperform the other state-of-the-art algorithms.

Self-Configuring (1 + 1)-Evolutionary Algorithm for the Continuous p-Median Problem with Agglomerative Mutation

The continuous p-median problem (CPMP) is one of the most popular and widely used models in location theory that minimizes the sum of distances from known demand points to the sought points called centers or medians. This NP-hard location problem is also useful for clustering (automatic grouping). In this case, sought points are considered as cluster centers. Unlike similar k-means model, p-median clustering is less sensitive to noisy data and appearance of the outliers (separately located demand points that do not belong to any cluster). Local search algorithms including Variable Neighborhood Search as well as evolutionary algorithms demonstrate rather precise results. Various algorithms based on the use of greedy agglomerative procedures are capable of obtaining very accurate results that are difficult to improve on with other methods. The computational complexity of such procedures limits their use for large problems, although computations on massively parallel systems significantly expand their capabilities. In addition, the efficiency of agglomerative procedures is highly dependent on the setting of their parameters. For the majority of practically important p-median problems, one can choose a very efficient algorithm based on the agglomerative procedures. However, the parameters of such algorithms, which ensure their high efficiency, are difficult to predict. We introduce the concept of the AGGLr neighborhood based on the application of the agglomerative procedure, and investigate the search efficiency in such a neighborhood depending on its parameter r. Using the similarities between local search algorithms and (1 + 1)-evolutionary algorithms, as well as the ability of the latter to adapt their search parameters, we propose a new algorithm based on a greedy agglomerative procedure with the automatically tuned parameter r. Our new algorithm does not require preliminary tuning of the parameter r of the agglomerative procedure, adjusting this parameter online, thus representing a more versatile computational tool. The advantages of the new algorithm are shown experimentally on problems with a data volume of up to 2,000,000 demand points.

Targeted LC-ESI-MS2 characterization of human milk oligosaccharide diversity at 6 to 16 weeks post-partum reveals clear staging effects and distinctive milk groups

Many molecular components in human milk (HM), such as human milk oligosaccharides (HMOs), assist in the healthy development of infants. It has been hypothesized that the functional benefits of HM may be highly dependent on the abundance and individual fine structures of contained HMOs and that distinctive HM groups can be defined by their HMO profiles. However, the structural diversity and abundances of individual HMOs may also vary between milk donors and at different stages of lactations. Improvements in efficiency and selectivity of quantitative HMO analysis are essential to further expand our understanding about the impact of HMO variations on healthy early life development. Hence, we applied here a targeted, highly selective, and semi-quantitative LC-ESI-MS2 approach by analyzing 2 × 30 mature human milk samples collected at 6 and 16 weeks post-partum. The analytical approach covered the most abundant HMOs up to hexasaccharides and, for the first time, also assigned blood group A and B tetrasaccharides. Principal component analysis (PCA) was employed and allowed for automatic grouping and assignment of human milk samples to four human milk groups which are related to the maternal Secretor (Se) and Lewis (Le) genotypes. We found that HMO diversity varied significantly between these four HM groups. Variations were driven by HMOs being either dependent or independent of maternal genetic Se and Le status. We found preliminary evidence for an additional HM subgroup within the Se- and Le-positive HM group I. Furthermore, the abundances of 6 distinct HMO structures (including 6′-SL and 3-FL) changed significantly with progression of lactation.