abstract: Multi-label learning, which deals with data associated with multiple labels simultaneously, is ubiquitous in real-world applications. To overcome the curse of dimensionality in multi-label learning, in this thesis I study multi-label dimensionality reduction, which extracts a small number of features by removing the irrelevant, redundant, and noisy information while considering the correlation among different labels in multi-label learning. Specifically, I propose Hypergraph Spectral Learning (HSL) to perform dimensionality reduction for multi-label data by exploiting correlations among different labels using a hypergraph. The regularization effect on the classical dimensionality reduction algorithm known as Canonical Correlation Analysis (CCA) is elucidated in this thesis. The relationship between CCA and Orthonormalized Partial Least Squares (OPLS) is also investigated. To perform dimensionality reduction efficiently for large-scale problems, two efficient implementations are proposed for a class of dimensionality reduction algorithms, including canonical correlation analysis, orthonormalized partial least squares, linear discriminant analysis, and hypergraph spectral learning. The first approach is a direct least squares approach which allows the use of different regularization penalties, but is applicable under a certain assumption; the second one is a two-stage approach which can be applied in the regularization setting without any assumption. Furthermore, an online implementation for the same class of dimensionality reduction algorithms is proposed when the data comes sequentially. A Matlab toolbox for multi-label dimensionality reduction has been developed and released. The proposed algorithms have been applied successfully in the Drosophila gene expression pattern image annotation. The experimental results on some benchmark data sets in multi-label learning also demonstrate the effectiveness and efficiency of the proposed algorithms. / Dissertation/Thesis / Ph.D. Computer Science 2011

The aim of this diploma paper is to evaluate the applicability and usability of modern evaluation methods for the financial situation in the company, focused on indicators EVA, MVA, and the cost of capital. Firstly, there are some basic terms defined in the theoretical part. The methodological part describes single the steps of calculations that were done in order to find out the answers for these hypothetical assumptions: 1.Evaluate whether it is possible to substitute difficult to detect characteristics of modern EVA and MVA indicators by different, easier indicator which will have at least the same explanatory power. 2.EVA indicator is able to predict the future development of a business as well as traditional predication models. These hypotheses were tested on a sample of 100 Czech firms in the construction industry. The data source was the database Albertina, which was purchased through a grant GAJU 053/2016/S. In the practical part are introduced results and its interpretations. The thesis conclusion is focused on the evaluation of particular hypotheses. The analysis proved: irreplaceability of indicators EVA and MVA in the success evaluation of company in certain year and inability of indicators EVA, MVA and predicting models as well to predict the future evolution

Orientador: Robson Pederiva / Dissertação (mestrado) -Universidade Estadual de Campinas, Faculdade de Engenharia Mecanica / Made available in DSpace on 2018-08-12T18:23:14Z (GMT). No. of bitstreams: 1 Sanches_FabioDalmazzo_M.pdf: 633170 bytes, checksum: a1620e7b33649f4441407be8d0f78500 (MD5) Previous issue date: 2008 / Resumo: Este trabalho apresenta uma metodologia de identificação dos parâmetros de rigidez e amortecimento dos mancais de rolamento de um sistema mecânico rotativo. O método proposto é baseado na equação matricial de Ljapunov e na representação do sistema na forma de espaço de estados. Através da definição de correlações, inserida na equação matricial, monta-se um estimador que relaciona os parâmetros físicos do sistema com as matrizes de correlações das variáveis medidas no domínio do tempo, não sendo necessário o conhecimento da excitação. O objetivo do trabalho é fazer um estudo teórico da aplicação dessa metodologia de identificação em sistemas reais. São feitas simulações considerando-se um sistema rotativo de vinte graus de liberdade excitado aleatoriamente, caso estático, e por desbalanceamento em diversas freqüências de rotação e de características de mancal. Os resultados numéricos demonstram que o método proposto é robusto e viável, podendo ser aplicado na identificação de uma máquina real / Abstract: This work presents an identification methodology of stiffness and damping parameters of rolling bearings in a rotor-bearing system. The proposed method is based on Ljapunov matrix equation and state space representation. Through the definition of correlations used in matrix equation, it is possible to generate an algorithm that relates system physical parameters to the correlation matrixes of the measured variables in time domain without the knowledge of the external input forces. The purpose of this work is to make a theoretical study of the viability of this methodology used in real systems. Simulations were performed considering a rotating system with twenty degrees of freedom excited by random forces, system at rest, and by unbalance forces considering several rotation frequencies and bearing characteristics. The numerical results show the proposed method is feasible and it can be used to identify real machines / Mestrado / Mecanica dos Sólidos e Projeto Mecanico / Mestre em Engenharia Mecânica

Neste trabalho considera-se uma reparametrização no modelo log-gama generalizado para a inclusão de dados com sobreviventes de longa duração. Os modelos tentam estimar separadamente os efeitos das covariáveis na aceleração ou desaceleração no tempo e na fração de sobreviventes que é a proporção da população para o qual o evento não ocorre. A função logística é usada para o modelo de regressão com fração de cura. Os parâmetros do modelo, serão estimados através do método de máxima verossimilhança. Alguns métodos de influência, como a influência local e a influência local total de um indivíduo, serão introduzidos, calculados, analisados e discutidos. Finalmente, um conjunto de dados médicos será analisado sob o modelo log-gama generalizado com fração de cura. Uma análise de resíduos será executada para verificar a qualidade de ajuste do modelo. / In this work the generalized log-gama model is modified for possibility that long-term survivors are present in the data . The models attempt to estimate separately the effects of covariates on the accelaration/decelaration of the timing of a given event and surviving fraction; that is, the proportion of the population for which the event never occurs. The logistic function is used for the regression model of the surviving fraction. Inference for the model parameters is considered via maximum likelihood. Some influence methods, such as the local influence, total local influence of an individual are derived, analyzed and discussed. Finally, a data set from the medical area is analyzed under log-gama generalized mixture model. A residual analysis is performed in order to select an appropriate model.

This is a short summary of a recent survey [FR03] focusing on the observed evidence, that Internet connectivity is positively correlated with spread of democracy at high levels of significance. The results of multivariate correlation analysis and probabilities regression estimate models are based on the combined analysis of mid - 1991's, to 2001 data series of the Eurostat's and US Census Bureau, the World Bank, and OECD's statistical data service which track the growth of information technology and rating of freedom and democracy worldwide.(author's abstract)

My Diploma thesis is focused on economic evaluation of gratuitous goods, especially on hedonic price method. The core of this paper is theoretical description and practical usage of HPM. Main theme is "Influence of plantscape on flat price.

Diploma thesis tries to analyze with complexity the influence of tourism on Spanish economy. The importance of the industry for the current economic growth is shown by different tourism performance indicators. Apart from the recent and current state of the tourism industry in Spain, performance of other sectors of the national economy are examined. An important part of the thesis is the correlation analysis, whose objective is to quantify the importance of tourism in Spanish economy and identify the correlation.

There were two major goals of this study - to examine validity of scores for the Boholst Life Position Scale and to examine potential associations between life positions and affective traits. Two hundred seventy-seven students enrolled in undergraduate psychology classes at a large university volunteered for the study. Concurrent validity of scores for the life position scale was supported based on two compared instruments. Pearson product-moment correlations for the comparisons were -.765 and .617, both statistically significant at the p < .001 level. Factor analysis demonstrated that the scale could accurately be conceptualized as consisting of two factors - an "I" factor and a "You" factor. MANOVA, ANOVA, multiple linear regression, and canonical correlation analysis were used to examine associations between life positions and the affective traits of angry, sad, glad, social anxiety, loneliness, and satisfaction with life. Subjects were catagorized into four groups representing their life position: "I'm OK, you're OK," "I'm OK, you're not OK," "I'm not OK, you're OK," and "I'm not OK, you're not OK." A MANOVA employing life position as the independent variable with four levels and the six affective traits as the dependent variables demonstrated statistical significance (p < .001 level) and h2 was .505. All six separate ANOVAs, with life position as the independent variable and each separate affective trait as the dependent variable, revealed statistical significance (p < .001) and h2 varied from a high of .396 for the sadness variable to a low of .116 for social anxiety. Six separate multiple linear regression equations using two independent variables, a measure of self-esteem and a measure of the perceived OK-ness of others, and each separate affective trait as the dependent variable, showed statistical significance (p < .001). The average Adjusted R2 was .475. Both canonical correlation functions were statistically significant (Rc12 = .77 and Rc22 = .21). In summary, life positions were strongly associated with specific affective traits.

This dissertation: (a) investigated the degree to which the squared canonical correlation coefficient is biased in multivariate nonnormal distributions and (b) identified formulae that adjust the squared canonical correlation coefficient (Rc2) such that it most closely approximates the true population effect under normal and nonnormal data conditions. Five conditions were manipulated in a fully-crossed design to determine the degree of bias associated with Rc2: distribution shape, variable sets, sample size to variable ratios, and within- and between-set correlations. Very few of the condition combinations produced acceptable amounts of bias in Rc2, but those that did were all found with first function results. The sample size to variable ratio (n:v)was determined to have the greatest impact on the bias associated with the Rc2 for the first, second, and third functions. The variable set condition also affected the accuracy of Rc2, but for the second and third functions only. The kurtosis levels of the marginal distributions (b2), and the between- and within-set correlations demonstrated little or no impact on the bias associated with Rc2. Therefore, it is recommended that researchers use n:v ratios of at least 10:1 in canonical analyses, although greater n:v ratios have the potential to produce even less bias. Furthermore,because it was determined that b2 did not impact the accuracy of Rc2, one can be somewhat confident that, with marginal distributions possessing homogenous kurtosis levels ranging anywhere from -1 to 8, Rc2 will likely be as accurate as that resulting from a normal distribution. Because the majority of Rc2 estimates were extremely biased, it is recommended that all Rc2 effects, regardless of which function from which they result, be adjusted using an appropriate adjustment formula. If no rationale exists for the use of another formula, the Rozeboom-2 would likely be a safe choice given that it produced the greatest number of unbiased Rc2 estimates for the greatest number of condition combinations in this study.

Soybean [Glycine max (L.) Merr.] is the world’s most widely grown protein/oilseed crop and provides about 70% of global protein meal and 53% of vegetable oil in the United States. Soybean seed oil contains five major fatty acids, from which palmitic acid and stearic acid are two saturated fatty acids, oleic acid improves oxidative stability and linolenic acid is an essential fatty acid for human health. Soybean seed protein and oil are two important quality indices for soybean germplasm breeding. Soluble carbohydrates present in soybean meal provide metabolizable energy in livestock feed. To develop soybean germplasm with improved seed composition traits, it is important to discover novel source of seed fatty acid, protein, and carbohydrates traits. This dissertation aims to develop novel functional genomic technology coupled with an integrated approach for facilitating molecular soybean breeding. In this study, the first objective is to develop a high-throughput TILLING (Targeting Induced Local Lesions IN Genomes) by Target Capture Sequencing (TbyTCS) technology to improve the efficiency of discovering mutations in soybean. The robustness of this technology underlies the high yield of true mutations in genes controlling complex traits in soybean. Soybean mutagenized lines with modified fatty acids composition have been successfully developed to meet the different needs of end users. Altered fatty acids phenotypes have been associated with induced mutations in 3-ketoacyl-acyl carrier protein (ACP) synthase II (GmKASII), Delta-9-stearoyl-acyl carrier protein desaturase (GmSACPD), omega-6 fatty acid desaturase 2 (GmFAD2), and omega-3 fatty acid desaturase (GmFAD3) genes identified through TbyTCS. The second objective is to characterize the soybean acyl-ACP thioesterase gene family through a comprehensive analysis. The additional members have been discovered belonging to 16:0-ACP fatty acid thioesterase (GmFATB) gene family. The mutations at oleoyl-ACP fatty acid thioesterase (GmFATA1A) have been revealed to result in the high seed oleic acid content. The novel alleles of GmFATB genes have also been identified to confer low palmitic acid and high oleic acid phenotypes in soybean seeds. The third objective is to assess the phenotypic variations and correlation among seed composition traits in mutagenized soybean populations. Correlation analyses have been conducted among soybean carbohydrates, protein, and oil content of soybean mutagenized populations and germplasm lines. Chemical mutagenesis played an essential role in soybean breeding to generate novel and desired seed composition traits.