| |
|
Řízení IS/ICT se zaměřením na sourcing služeb informačního systému
Šebesta, Michal ; Voříšek, Jiří (advisor) ; Havlíček, Zdeněk (referee) ; Příklenk, Oldřich (referee) ; Král, Jaroslav (referee)
Research on outsourcing has been around for several decades, while recent evolution in the information systems discipline towards ICT service commoditization significantly changes the context of decision-making. Services that are available on-demand via the Internet allow organizations implementing functions they demand in a fraction of time. This trend represents a chance for organizations seeking to use advanced ICT services without a need of major investments. Problem is the current lack of guidelines and tools for managing ICT services and their outsourcing. Given the trends on the ICT service market, it is expected that much of the IT management in the future will encompass the ICT services and utilize service-level structures. Methods currently available are either too broad or encompass only small part of the whole problem. Ad-hoc or unsound decisions in this area might cause major complications in terms of quality, usability, integration, and consequently influence total cost of organizational IT. Organizations need to either revise existing models or propose and implement completely new models to manage their IS/ICT. This thesis deals with the management of IS/ICT with focus on the ICT services outsourcing. It discusses available sourcing models in the literature and links them to the various interconnected areas. Based on these areas, it presents an integrated view on IT outsourcing strategies. Most importantly the thesis proposes an original concept for decision-making about outsourcing of ICT services named the SOURCER framework. This approach utilizes the presented outsourcing strategies, and introduces a complex methodology and decision-making criteria that will assist organizations with selection of ICT services in order to maintain and manage a most suitable ICT service portfolio. The decision-making is based on four essential viewpoints: function, costs, time, and quality. These viewpoints are discussed, individually analyzed, and serve as a basis for further research. The whole framework is developed and validated according to Design Science Research Methodology (DSRM). Individual components are evaluated using a survey among a group of selected IT managers. Proof of concept is then established by a case study on framework use in a real organization. This case study covers strategy specification, business--IT alignment, specifying service architecture and its interconnections, outsourcing, and management of the ICT service portfolio.
|
|
Clustering and regression analysis of micro panel data
Sobíšek, Lukáš ; Pecáková, Iva (advisor) ; Komárek, Arnošt (referee) ; Brabec, Marek (referee)
The main purpose of panel studies is to analyze changes in values of studied variables over time. In micro panel research, a large number of elements are periodically observed within the relatively short time period of just a few years. Moreover, the number of repeated measurements is small. This dissertation deals with contemporary approaches to the regression and the clustering analysis of micro panel data. One of the approaches to the micro panel analysis is to use multivariate statistical models originally designed for crosssectional data and modify them in order to take into account the within-subject correlation. The thesis summarizes available tools for the regression analysis of micro panel data. The known and currently used linear mixed effects models for a normally distributed dependent variable are recapitulated. Besides that, new approaches for analysis of a response variable with other than normal distribution are presented. These approaches include the generalized marginal linear model, the generalized linear mixed effects model and the Bayesian modelling approach. In addition to describing the aforementioned models, the paper also includes a brief overview of their implementation in the R software. The difficulty with the regression models adjusted for micro panel data is the ambiguity of their parameters estimation. This thesis proposes a way to improve the estimations through the cluster analysis. For this reason, the thesis also contains a description of methods of the cluster analysis of micro panel data. Because supply of the methods is limited, the main goal of this paper is to devise its own two-step approach for clustering micro panel data. In the first step, the panel data are transformed into a static form using a set of proposed characteristics of dynamics. These characteristics represent different features of time course of the observed variables. In the second step, the elements are clustered by conventional spatial clustering techniques (agglomerative clustering and the C-means partitioning). The clustering is based on a dissimilarity matrix of the values of clustering variables calculated in the first step. Another goal of this paper is to find out whether the suggested procedure leads to an improvement in quality of the regression models for this type of data. By means of a simulation study, the procedure drafted herein is compared to the procedure applied in the kml package of the R software, as well as to the clustering characteristics proposed by Urso (2004). The simulation study demonstrated better results of the proposed combination of clustering variables as compared to the other combinations currently used. A corresponding script written in the R-language represents another benefit of this paper. It is available on the attached CD and it can be used for analyses of readers own micro panel data.
|
|
Míry podobnosti pro nominální data v hierarchickém shlukování
Šulc, Zdeněk ; Řezanková, Hana (advisor) ; Šimůnek, Milan (referee) ; Žambochová, Marta (referee)
This dissertation thesis deals with similarity measures for nominal data in hierarchical clustering, which can cope with variables with more than two categories, and which aspire to replace the simple matching approach standardly used in this area. These similarity measures take into account additional characteristics of a dataset, such as frequency distribution of categories or number of categories of a given variable. The thesis recognizes three main aims. The first one is an examination and clustering performance evaluation of selected similarity measures for nominal data in hierarchical clustering of objects and variables. To achieve this goal, four experiments dealing both with the object and variable clustering were performed. They examine the clustering quality of the examined similarity measures for nominal data in comparison with the commonly used similarity measures using a binary transformation, and moreover, with several alternative methods for nominal data clustering. The comparison and evaluation are performed on real and generated datasets. Outputs of these experiments lead to knowledge, which similarity measures can generally be used, which ones perform well in a particular situation, and which ones are not recommended to use for an object or variable clustering. The second aim is to propose a theory-based similarity measure, evaluate its properties, and compare it with the other examined similarity measures. Based on this aim, two novel similarity measures, Variable Entropy and Variable Mutability are proposed; especially, the former one performs very well in datasets with a lower number of variables. The third aim of this thesis is to provide a convenient software implementation based on the examined similarity measures for nominal data, which covers the whole clustering process from a computation of a proximity matrix to evaluation of resulting clusters. This goal was also achieved by creating the nomclust package for the software R, which covers this issue, and which is freely available.
|
|
Possibilities of Big Data use for Competitive Intelligence
Verníček, Marek ; Molnár, Zdeněk (advisor) ; Šperková, Lucie (referee)
The main purpose of this thesis is to investigate the use of Big Data for the methods and procedures of Competitive Intelligence. Among the goals of the work is a toolkit for small and large businesses which is supposed to support their work with the whole process of Big Data work. Another goal is to design an effective solution of processing Big Data to gain a competitive advantage in business. The theoretical part of the work processes available scientific literature in the Czech Republic and abroad as well as describes the current state of Competitive Intelligence, and Big Data as one of its possible sources. Subsequently, the work deals with the characteristics of Big Data, the differences from working with common data, the need for a thorough preparation and Big Data applicability for the methods of Competitive Intelligence. The practical part is focused on analysis of Big Data tools available in the market with regard to the whole process from data collection to the analysis report preparation and integration of the entire solution into an automated state. The outcome of this part is the Big Data software toolkit for small and large businesses based on their budget. The final part of the work is devoted to the classification of the most promising business areas, which can benefit from the use of Big Data the most in order to gain competitive advantages and proposes the most effective solution of working with Big Data. Among other benefits of this work are expansion of the range of resources for Competitive Intelligence and in-depth analysis of possibilities of Big Data usage, designed to help professionals make use of this hitherto untapped potential to improve market position, gain new customers and strengthen the existing user base.
|
|
The GIS support to measures on the ground in case of leakage of liquid pollutant on the road
Kolejka, Jaromír ; Rapant, P. ; Zapletalová, Jana
Accidents on roads associated with the leakage of hazardous substances are one of the major challenges encountered by disaster management. Because of the impossibility of predicting the place and time of the event, then it is necessary in the event of such an accident to proceed in quick succession of steps. They are designed primarily to protect human life and health, and then to minimize to property and environment damage. The paper describes the response to this event using GIS tools and generally available geodata. The simulated accident on the D1 highway near Ostrava is applied as a demonstration example.
|
|
Modelling, parameter estimation, optimisation and control of transport and reaction processes in bioreactors.
ŠTUMBAUER, Václav
With the significant potential of microalgae as a major biofuel source of the future, a considerable scientific attention is attracted towards the field of biotechnology and bioprocess engineering. Nevertheless the current photobioreactor (PBR) design methods are still too empirical. With this work I would like to promote the idea of designing a production system, such as a PBR, completely \emph{in silico}, thus allowing for the in silico optimization and optimal control determination. The thesis deals with the PBR modeling and simulation. It addresses two crucial issues in the current state-of-the-art PBR modeling. The first issue relevant to the deficiency of the currently available models - the incorrect or insufficient treatment of either the transport process modeling, the reaction modeling or the coupling between these two models. A correct treatment of both the transport and the reaction phenomena is proposed in the thesis - in the form of a unified modeling framework consisting of three interconnected parts - (i) the state system, (ii) the fluid-dynamic model and (iii) optimal control determination. The proposed model structure allows prediction of the PBR performance with respect to the modelled PBR size, geometry, operating conditions or a particular microalgae strain. The proposed unified modeling approach is applied to the case of the Couette-Taylor photobioreactor (CTBR) where it is used for the optimal control solution. The PBR represents a complex multiscale problem and especially in the case of the production scale systems, the associated computational costs are paramount. This is the second crucial issue addressed in the thesis. With respect to the computational complexity, the fluid dynamics simulation is the most costly part of the PBR simulation. To model the fluid flow with the classical CFD (Computational Fluid Dynamics) methods inside a production scale PBR leads to an enormous grid size. This usually requires a parallel implementation of the solver but in the parallelization of the classical methods lies another relevant issue - that of the amount of data the individual nodes must interchange with each other. The thesis addresses the performance relevant issues by proposing and evaluation alternative approaches to the fluid flow simulation. These approaches are more suitable to the parallel implementation than the classical methods because of their rather local character in comparison to the classical methods - namely the Lattice Boltzmann Method (LBM) for fluid flow, which is the primary focus of the thesis in this regard and alternatively also the discrete random walk based method (DRW). As the outcome of the thesis I have developed and validated a new Lagrangian general modeling approach to the transport and reaction processes in PBR - a framework based on the Lattice Boltzmann method (LBM) and the model of the Photosynthetic Factory (PSF) that models correctly the transport and reaction processes and their coupling. Further I have implemented a software prototype based on the proposed modeling approach and validated this prototype on the case of the Coutte-Taylor PBR. I have also demonstrated that the modeling approach has a significant potential from the computational costs point of view by implementing and validating the software prototype on the parallel architecture of CUDA (Compute Unified Device Architecture). The current parallel implementation is approximately 20 times faster than the unparallized one and decreases thus significantly the iteration cycle of the PBR design process.
|
|
Methods for class prediction with high-dimensional gene expression data
Šilhavá, Jana ; Matula, Petr (referee) ; Železný, Filip (referee) ; Smrž, Pavel (advisor)
Dizertační práce se zabývá predikcí vysokodimenzionálních dat genových expresí. Množství dostupných genomických dat významně vzrostlo v průběhu posledního desetiletí. Kombinování dat genových expresí s dalšími daty nachází uplatnění v mnoha oblastech. Například v klinickém řízení rakoviny (clinical cancer management) může přispět k přesnějšímu určení prognózy nemocí. Hlavní část této dizertační práce je zaměřena na kombinování dat genových expresí a klinických dat. Používáme logistické regresní modely vytvořené prostřednictvím různých regularizačních technik. Generalizované lineární modely umožňují kombinování modelů s různou strukturou dat. V dizertační práci je ukázáno, že kombinování modelu dat genových expresí a klinických dat může vést ke zpřesnění výsledku predikce oproti vytvoření modelu pouze z dat genových expresí nebo klinických dat. Navrhované postupy přitom nejsou výpočetně náročné. Testování je provedeno nejprve se simulovanými datovými sadami v různých nastaveních a následně s~reálnými srovnávacími daty. Také se zde zabýváme určením přídavné hodnoty microarray dat. Dizertační práce obsahuje porovnání příznaků vybraných pomocí klasifikátoru genových expresí na pěti různých sadách dat týkajících se rakoviny prsu. Navrhujeme také postup výběru příznaků, který kombinuje data genových expresí a znalosti z genových ontologií.
|
|
Crystallographic study of the iron-regulated outer membrane lipoprotein (FrpD) from Neisseria meningitidis
SVIRIDOVA, Ekaterina
Neisseria meningitidis (N. meningitidis) is a Gram-negative commensal bacterium colonizing nasopharynx of about 10 % of healthy individuals, which can cause invasive diseases, such sepsis and meningitis, upon occasional penetration into bloodstream. Pathogenesis of N. meningitidis appears to be directly related to conditions of limited iron availability. Under these conditions two proteins of unknown function: FrpC and FrpD, are synthesized. FrpD is a highly conserved lipoprotein of N. meningitidis anchored to the bacterial outer membrane. It is known that FrpD tightly binds the FrpC protein, which belongs to the Repeat-in-Toxin (RTX) protein family and may act as bacterial exotoxin. However, the mechanism of FrpD-FrpC interaction and the exact function of this complex are unknown due to the absence of structural information on these proteins. Therefore, we set out to determine the structure of FrpD and provide insights into its interaction mechanism with FrpC and structure-functional relationships of these two proteins. We determined the first crystal and solution structures of the FrpD protein. We found that atomic structures of FrpD reveal a novel protein fold. We uncovered the structure-function relationships underlying the mechanism of interaction between the FrpD and FrpC proteins and tested the putative function of the FrpD-FrpC1-414 complex in vitro. Finally, we proposed the putative function of the FrpD-FrpC1-414 complex as a new minor adhesin of N. meningitidis, which mediates the bacterial adhesion to the host epithelial cells and facilitate the colonization. Our work constitutes the first step in clarifying the molecular basis of the FrpD-FrpC interaction and sets the base for further investigation of the role of FrpD and FrpC in the virulence mechanism of N. meningitidis.
|
|
Is it possible in clinical practice to perform selection of unrelated donors based on KIR genotypes for AML patients?
FRYČOVÁ, Michaela
Acute myeloid leukemia (AML) is an aggressive malignant disease, during which is for most of the patients only possible treatment the curative allogeneic stem cell transplantation. Besides reaction of the graft against the host is a fundamental limiting factor of the successful transplantation the relapse of the disease. According to several recent published studies, the results of transplantation in patients with AML may be influenced except the HLA genes by others so-called non - HLA genes. Especially there is mounting evidence influence of the donors KIR genes (Killer -cell Immunoglobulin - like receptors) in protection against the relapse after transplantation. HLA and KIR genes are coded on different chromosomes (HLA- sixth chromosome and KIR chromosome 19), therefore are segregated independently and HLA identical donors with recipients usually have different compositions of the KIR genes. Cooley et al. (2010) demonstrated that the specific motifs composition of centromeric and telomeric B haplotypes of KIR genes helps to protect against relapse and increases the chances of complete cure AML. In cases where there are multiple HLA identical unrelated donors (UD) then logically the composition KIR genes by the individual donor could be a criterion in selecting the most appropriate donor, therefore, the one with the greatest potential to protect over the relapse . Based on this study and other data the genetic screening of KIR was started with potential donors, if it was possible to choose from several 10/10 or 10/09 HLA identical UD for the patient. Genotyping was performed by PCR-SSP methodology using commercially available kits. It was performed gene classification 160 preferably identical HLA donors for 55 selected patients with AML. The presence of KIR haplotypes A and B as well as their combinations was determined from the type and number of the KIR genes. All genotypes were entered into the calculator, which allows you to enter up to five potential donors and obtain their assigned into one of three categories according to content KIR B. Groups , "neutral" , "better" , "best" , refer to the appropriate protection against relapse. KIR gene classification in the search for donors revealed 43 donors with AA haplotypes, 90 donors with AB haplotypes and 27 donors with BB haplotypes . After assigning state of the presence KIR B was discovered 107 " neutral " donors , 35 "better " donors and the 18 "best " donors . At 40 (~ 73 %) patients were available donors with the different states of the presence of KIR B. These patients represent a group of patients where the selection criterion of the presence B KIR gene at the donor could be used. We confirmed that the additional selection of HLA-matched unrelated donor on the basis of the content B of KIR genes is feasible. Selection such donor for transplantation may improve the outcome of patients with AML.
|
guest ::
