The thesis deals with the comparison of conventional and heuristic methods in data mining used for binary classification. In the theoretical part, four different models are described. Model classification is demonstrated on simple examples. In the practical part, models are compared on real data. This part also consists of data cleaning, outliers removal, two different transformations and dimension reduction. In the last part methods used to quality testing of models are described.

The thesis discusses in detail some concepts of speech and prosody that can contribute to build a speech corpus for the speaker segmentation purpose. Moreover, the Elan multimedia annotator used for labeling is described. The theoretical part highlights some frequently used speech features such as MFCC, PLP and LPC and deals with currently most popular speech segmentation methods. Some classification algorithms are also mentioned. The practical part describes implementation of Bayesian information criterium algorithm in system for automatic speaker segmentation. For classification of speaker change point in speech, were used different speech features. The results of tests were evaluated by the graphic method of receiver operating characteristic (ROC) and his quantitative indices. As the best speech features for this system were provided MFCC and HFCC.

Indiana University-Purdue University Indianapolis (IUPUI) / Parkinson's disease (PD) is a progressive neurodegenerative disorder that affects over 10 million people. Treatments for PD are limited to symptom mitigation with no means of stopping or slowing disease progression. Mutations within the protein leucine- rich repeat kinase 2 (LRRK2) are the most common cause of familial PD and are indistinguishable from the more common sporadic cases. Identifying molecules capable of modulating LRRK2 GTPase activity may serve as the foundation for future development of novel PD therapeutics. We recently discovered that the G-domain (ROC) of LRRK2 is capable of transitioning between monomer and dimer form in solution upon GTP/GDP binding. R1441C/G/H pathogenic mutations were demonstrated to alter this dynamic shifting toward a monomeric ROC conformation while decreasing GTPase activity. Using our ROC dimeric crystal structure, we strategically introduced disulfide bonds to generate locked monomer and locked dimer states. Monomeric ROC was shown to increase GTPase activity while the dimeric form decreased activity. Solvent mapping performed using the dimeric ROC crystal structure and a homology model of the ROC monomer revealed a binding hotspot at the ROC dimeric interface and adjacent to the R1441 residue in the monomeric model. In this study our goal was to identify more compounds capable of influencing GTPase activity. We performed high throughput screening of ROC against two compound libraries (LOPAC1280 and ChemBridge 50K) in a GTP binding assay. Twenty-three hits were identified and four compounds were further investigated in dose-response experiments. 3,4-Methylenedioxy-beta nitrostyrene (MNS) was demonstrated to decrease GTP binding and inhibit GTPase activity (IC50=23.92μM) while the compound N-phenylanthranilic acid increased GTP binding (EC50=4.969μM) and decreased GTPase activity. Identification of these compounds is the first step in the development of a novel PD therapeutic targeting the G-domain of LRRK2.

Bivariate and multivariate statistical analyses were used to predict the spatial distribution of landslides in the Cuyahoga River watershed, northeastern Ohio, U.S.A. The relationship between landslides and various instability factors contributing to their occurrence was evaluated using a Geographic Information System (GIS) based investigation. A landslide inventory map was prepared using landslide locations identified from aerial photographs, field checks, and existing literature. Instability factors such as slope angle, soil type, soil erodibility, soil liquidity index, landcover pattern, precipitation, and proximity to stream, responsible for the occurrence of landslides, were imported as raster data layers in ArcGIS, and ranked using a numerical scale corresponding to the physical conditions of the region. In order to investigate the role of each instability factor in controlling the spatial distribution of landslides, both bivariate and multivariate models were used to analyze the digital dataset. The logistic regression approach was used in the multivariate model analysis. Both models helped produce landslide susceptibility maps and the suitability of each model was evaluated by the area under the curve method, and by comparing the maps with the known landslide locations. The multivariate logistic regression model was found to be the better model in predicting landslide susceptibility of this area. The logistic regression model produced a landslide susceptibility map at a scale of 1:24,000 that classified susceptibility into four categories: low, moderate, high, and very high. The results also indicated that slope angle, proximity to stream, soil erodibility, and soil type were statistically significant in controlling the slope movement.

The goal of this theses is to compare existing bankruptcy prediction models with its new modification unique for this work, which could perform better than its competition. Proposed model is logit-based and consists of the combination of variables used in Altman´s and Ohlson´s models. The final model is estimated for medium sized companies in EU which aren´t publicly traded. This model achieved prediction accuracy of 97,1% (97.4% for healthy and 91.1% for bankrupt compa-nies) on its original dataset. As expected, when verified on new dataset, the accu-racy dropped but still reaches 97.1% (99.3% for healthy and 37.7% for bankrupt companies). The model is compared with its competition (original and modified version of Ohlson´s and partially Altman´s models) and it is shown that it has higher prediction accuracy.

京都大学 / 0048 / 新制・課程博士 / 博士(社会健康医学) / 甲第19969号 / 社医博第74号 / 新制||社医||9(附属図書館) / 33065 / 京都大学大学院医学研究科社会健康医学系専攻 / (主査)教授 小池 薫, 教授 福田 和彦, 教授 木村 剛 / 学位規則第4条第1項該当 / Doctor of Public Health / Kyoto University / DFAM

No description available.

This document explains the use of different metrics involved with record linkage. There are two forms of record linkage: deterministic and probabilistic. We will focus on probabilistic record linkage used in merging and updating two databases. Record pairs will be compared using character-based and phonetic-based similarity metrics to determine at what level they match. Performance measures are then calculated and Receiver Operating Characteristic (ROC) curves are formed. Finally, an economic model is applied that returns the optimal tolerance level two databases should use to determine a record pair match in order to maximize profit.

Finite mixture models have had a profound impact on the history of statistics, contributing to modelling heterogeneous populations, generalizing distributional assumptions, and lately, presenting a convenient framework for classification and clustering. A novel approach, via Gaussian mixture distribution, is introduced for modelling receiver operating characteristic curves. The absence of a closed-form for a functional form leads to employing the Monte Carlo method. This approach performs excellently compared to the existing methods when applied to real data. In practice, the data are often non-normal, atypical, or skewed. It is apparent that non-Gaussian distributions be introduced in order to better fit these data. Two non-Gaussian mixtures, i.e., t distribution and skew t distribution, are proposed and applied to real data. A novel mixture is presented to cluster spatial and temporal data. The proposed model defines each mixture component as a mixture of autoregressive polynomial with logistic links. The new model performs significantly better compared to the most well known model-based clustering techniques when applied to real data. / Thesis / Doctor of Philosophy (PhD)

No description available.