|
|
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.
|
|
|
Building predictive models
ZABLOUDIL, Jakub
This mater thesis is focused on building predictive models. Their fundamental task is to provide an early-warning system, giving information about potential enterprise bankruptcy. The main essence and aim of the thesis is to create multivariate classification models by using discriminant analysis and logistic regression. Emphasis is put on their predictive accuracy, which is assessed for period of three years before bankruptcy declaration. Attempts to optimize classification thresholds in order to increase the initial accuracy are also made. Evaluating classification reliability of several existing models and performing profile analysis assessing predictive ability of univariate ratios were accomplished as well.
|
|
|
Evolutionary Design of Simulator Based on Cellular Automata
Brigant, Vladimír ; Šperka, Svatopluk (referee) ; Mrnuštík, Michal (advisor)
This work describes concept of a cellular automata (CA) simulator, which is able to predict behaviour of a complex spatial system. This prediction is based on available training data and transition rule acquired from regression analysis powered by evolutionary algorithms. Two regression analysis methods (linear and logistic regression) are suggested, implemented and compared on urban growth prediction of Brno city.
|
|
|
Deep Learning for Image Recognition
Munzar, Milan ; Kolář, Martin (referee) ; Hradiš, Michal (advisor)
Neural networks are one of the state-of-the-art models for machine learning today. One may found them in autonomous robot systems, object and speech recognition, prediction and many others AI tasks. The thesis describes this model and its extension which is used in an object recognition. Then explains an application of a convolutional neural networks(CNNs) in an image recognition on Caltech101 and Cifar10 datasets. Using this exemplar application, the thesis discusses and measures efficiency of techniques used in CNNs. Results show that the convolutional networks without advanced extensions are able to reach a 80\% recognition accuracy on Cifar-10 and a 37\% accuracy on Caltech101.
|
|
|
Statistical Classification by means of generalized linear models
Sladká, Vladimíra ; Mrázková, Eva (referee) ; Michálek, Jaroslav (advisor)
The goal of this thesis is introduce the theory of generalized linear models, namely probit and logit model. This models are especially used for medical data processing. In our concrete case these mentioned models are applied to data file obtained in teaching hospital Brno. The aim is statically analyzed immune response of child patients in dependence of twelve selected types of genes and find out which combinations of these genes influence septic state of patients.
|
|
|
Statistical Models of Success of Various Techniques of Rugby Kicking
Vrbacká, Kateřina ; Votavová, Helena (referee) ; Bednář, Josef (advisor)
This bachelor thesis is dealing with the testing of statistical hypothesis and their practical use. We model the success of rugby kicking and analyze the dominant factors (ball position, kicking technique, player) and their interactions. We will use some mathematical terms such as chi-square test of independence and logistic regression. The final model will be processed by software MINITAB. The outcome from this thesis will be the exact description of this situation.
|
|
|
The Estimation of Probability of Default Using Logistic Regression
Chalupa, Tomáš ; Dlouhá, Zuzana (advisor) ; Formánek, Tomáš (referee)
The aim of this work is to develop a suitable model that estimates a probability of default of client's loan. As estimation method was used a logistic regression and a probit regression and two definitions of default, 60 and 90 days overdue. The work describes the method of construction, estimation and testing of scoring models and a structure of dataset, which was used in the practical part. Firstly, it was created a theoretical model that was later confronted with estimates. Estimated models were compared by described statistics as McFadden R^2, the ability to diversify was investigated by the Lorenz curve and by the Gini coefficient. It was found that the logistic and the probit regressions have almost the same results, and that 90 days is preferable definition of default than 60 days.
|
|
|
CHAID and logistic regression
Novák, Jaroslav ; Čabla, Adam (advisor) ; Matějka, Martin (referee)
The aim of this thesis is to introduce logistic regression and method dedicated to construction of decision trees called CHAID, compare these two methods with regard to interpretation of their outputs. In order to accomplish the set goals application of these methods will be presented on real dataset. Statistical software will be used to obtain outputs. The outputs will be interpreted and conclusions on their bases will be presented. This thesis will also introduce possibilities of interpretation of these outputs and pros and cons that are connected with them.
|
|
|
On-line Services in Hospitals
Venkrbcová, Eva ; Bína, Vladislav (advisor) ; Lešetický, Ondřej (referee)
The thesis is focused on mapping of selected on-line services in hospitals in Czech Re-public according to their availability. In the next step logistic regression model will be used, which will measure a range of particular on-line services according to their legal form, majority ownership, region in which health organization is located, to education achieved and gender of hospital director and lastly in accordance with the size of hospital from the viewpoint of bed capacity. This thesis will provide comprehensive overview of availability of on-line services in Czech hospitals and determine which factors significantly decreasing or increasing chance of their occurrence on hospital websites.
|
|
|
Data Mining and use of decision trees by creation of Scorecards
Straková, Kristýna ; Witzany, Jiří (advisor) ; Fičura, Milan (referee)
The thesis presents a comparison of several selected modeling methods used by financial institutions for (not exclusively) decision-making processes. First theoretical part describes well known modeling methods such as logistic regression, decision trees, neural networks, alternating decision trees and relatively new method called "Random forest". The practical part of thesis outlines some processes within financial institutions, in which selected modeling methods are used. On real data of two financial institutions logistic regression, decision trees and decision forest are compared which each other. Method of neural network is not included due to its complex interpretability. In conclusion, based on resulting models, thesis is trying to answers, whether logistic regression (method most widely used by financial institutions) remains most suitable.
|
guest ::
