Malgorzata Charytanowicz

Complete Gradient Clustering Algorithm for Features Analysis of X-Ray Images

Methods based on kernel density estimation have been successfully applied for various data mining tasks. Their natural interpretation together with suitable properties make them an attractive tool among others in clustering problems. In this paper, the Complete Gradient Clustering Algorithm has been used to investigate a real data set of grains. The wheat varieties, Kama, Rosa and Canadian, characterized by measurements of main grain geometric features obtained by X-ray technique, have been analyzed. The proposed algorithm is expected to be an effective tool for recognizing wheat varieties. A comparison between the clustering results obtained from this method and the classical k-means clustering algorithm shows positive practical features of the Complete Gradient Clustering Algorithm.

A complete gradient clustering algorithm formed with kernel estimators

A complete gradient clustering algorithm formed with kernel estimators The aim of this paper is to provide a gradient clustering algorithm in its complete form, suitable for direct use without requiring a deeper statistical knowledge. The values of all parameters are effectively calculated using optimizing procedures. Moreover, an illustrative analysis of the meaning of particular parameters is shown, followed by the effects resulting from possible modifications with respect to their primarily assigned optimal values. The proposed algorithm does not demand strict assumptions regarding the desired number of clusters, which allows the obtained number to be better suited to a real data structure. Moreover, a feature specific to it is the possibility to influence the proportion between the number of clusters in areas where data elements are dense as opposed to their sparse regions. Finally, the algorithm—by the detection of oneelement clusters—allows identifying atypical elements, which enables their elimination or possible designation to bigger clusters, thus increasing the homogeneity of the data set.

Th17 and Treg cells in adolescents with Graves' disease. Impact of treatment with methimazole on these cell subsets.

Abstract The role of T helper 17 (Th17) and T regulatory cells (Treg) in the pathogenesis of Graves’ disease (GD) remains uncertain. The influence of methimazole (MMI) on the human immune system is still poorly understood. The aim of the present research was to assess changes in the frequencies of peripheral blood Th17 and Treg cells during GD treatment in the group of teenagers. The frequencies of Th17 and Treg were measured by flow cytometry in 60 adolescents at the time of GD diagnosis and after achieving MMI-induced euthyreosis. The control group consisted of 20 healthy volunteers. Lower percentages and absolute counts of Treg cells were found in the study group before the treatment in comparison with healthy controls (p = 0.032 and p = 0.006, respectively). Treatment with MMI caused an increase in the percentages and absolute counts of Treg lymphocytes (p = 0.037 and p = 0.007). After the treatment, no clinically significant differences in Treg cells between GD patients and controls were found. Higher absolute counts of Th17 lymphocytes were found in hyperthyroid adolescents before the treatment initiation and after achieving euthyreosis than in healthy individuals (p = 0.0001 and p = 0.047). Treatment with MMI caused a significant decrease in the percentages and absolute counts of Th17 lymphocytes (p = 0.047 and p = 0.043). The present study demonstrates that both Th17 and Treg cells might play a role in the pathogenesis of GD. Increased percentage of Treg after MMI therapy seems a predictor of response to anti-hypertensive treatment as it is associated with the normalization of thyroid hormone levels.

The Complete Gradient Clustering Algorithm: properties in practical applications

The aim of this paper is to present a Complete Gradient Clustering Algorithm, its applicational aspects and properties, as well as to illustrate them with specific practical problems from the subject of bioinformatics (the categorization of grains for seed production), management (the design of a marketing support strategy for a mobile phone operator) and engineering (the synthesis of a fuzzy controller). The main property of the Complete Gradient Clustering Algorithm is that it does not require strict assumptions regarding the desired number of clusters, which allows to better suit its obtained number to a real data structure. In the basic version it is possible to provide a complete set of procedures for defining the values of all functions and parameters relying on the optimization criterions. It is also possible to point out parameters, the potential change which implies influence on the size of the number of clusters (while still not giving an exact number) and the proportion between their numbers in dense and sparse areas of data elements. Moreover, the Complete Gradient Clustering Algorithm can be used to identify and possibly eliminate atypical elements (outliers). These properties proved to be very useful in the presented applications and may also be functional in many other practical problems.

A complete gradient clustering algorithm

A gradient clustering algorithm, based on the nonparametric methodology of statistical kernel estimators, expanded to its complete form, enabling implementation without particular knowledge of the theoretical aspects or laborious research, is presented here. The possibilities of calculating tentative optimal parameter values, and then - based on illustrative interpretation - their potential changes, result in the proposed Complete Gradient Clustering Algorithm possessing many original and valuable, from an applicational point of view, properties. Above all the number of clusters is not arbitrarily imposed but fitted to a real data structure. It is also possible to increase the scale of the number (still avoiding arbitrary assumptions), as well as the proportion of clusters in areas of dense and sparse situation of data elements. The method is universal in character and can be applied to a wide range of practical problems, in particular from the bioinformatics, management and engineering fields.

Clustering using flower pollination algorithm and Calinski-Harabasz index

Task of clustering, that is data division into homogeneous groups represents one of the elementary problems of contemporary data mining. Cluster analysis can be approached through variety of methods based on statistical inference or heuristic techniques. Recently algorithms employing novel meta-heuristics are of special interest - as they can effectively tackle the problem under consideration which is known to be NP-hard. The paper studies the application of nature-inspired Flower Pollination Algorithm for clustering with internal measure of Calinski-Harabasz index being used as optimization criterion. Along with algorithms description its performance is being evaluated over a set of benchmark instances and compared with the one of well-known K-means procedure. It is concluded that the application of introduced technique brings very promising outcomes. The discussion of obtained results is followed by areas of possible improvements and plans for further research.

Fuzzy Models Synthesis with Kernel-Density-Based Clustering Algorithm

Data clustering constitutes at present a commonly used technique for extracting fuzzy system rules from experimental data. Detailed studies in the field have shown that using above-mentioned method results in significantly reduced structure of fuzzy identification system, maintaining at the same time its high modelling efficiency. In this paper a clustering algorithm, based on a kernel density gradient estimation procedure applied for fuzzy models synthesis, is presented. It consists of two stages: data elements relocation and their division into clusters. The method is automatic, unsupervised, and does not require any assumptions concerning the desired number of fuzzy rules. The results of experimental evaluation show that the algorithm under consideration achieves relatively high performance when compared to the standard techniques frequently applied in similar class of problems.

Clustering based on the Krill Herd Algorithm with selected validity measures

This paper describes a new approach to metaheuristic-based data clustering by means of Krill Herd Algorithm (KHA). In this work, KHA is used to find centres of the cluster groups. Moreover, the number of clusters is set up at the beginning of the procedure, and during the subsequent iterations of the optimization algorithm, particular solutions are evaluated by selected validity criteria. The proposed clustering algorithm has been numerically verified using twelve data sets taken from the UCI Machine Learning Repository. Additionally, all cases of clustering were compared with the most popular method of k-means, through the Rand Index being applied as a validity measure.

ASYMMETRICAL CONDITIONAL BAYES PARAMETER IDENTIFICATION FOR CONTROL ENGINEERING

One of the main problems in control engineering practice results from unavoidable errors in specifying parameters existing in the object model and the necessity to deal with the unwanted phenomena arising as a result. In this article, a Bayes methodology considering both asymmetrical and conditional aspects is applied for this purpose, with the application of kernel estimators methodology. Use of the Bayes rule enables minimum potential losses to be assumed, while the asymmetry of the occurring loss function also enables the inclusion of different results for under- and overestimation. A conditional approach allows researchers to obtain a more precise result thanks to using information entered as the fixed (e.g., current) values of conditioning factors of continuous and/or binary (also categorical) type. The nonparametric methodology of statistical kernel estimators frees the procedure from arbitrary assumptions concerning the forms of distributions characterizing both the parameter under investigation and conditioning factors. The generalizations introduced here also allow different relevance of particular random sample elements to be taken into account.

Nonparametric Regression for Analyzing Correlation between Medical Parameters

Chronic renal failure is associated with major biochemical and hematological derangements. These changes are often represented as linear functions of creatinine. The aim of the study is to analyze the correlation of hematologic parameters with creatinine. The sample population involved patients with renal insufficiency observed in the Stefan Kardynal Wyszynski Regional Specialists’ Hospital in Lublin (Poland). The method presented here is based on the theory of statistical kernel estimators, which frees it of assumptions in regard to the form of regression function.