Thesis (Ph. D.)--Tulane University, 1971. / Typescript. Words of the musical selections also printed as text (v. 2, leaves vi-xxix). Contains vita. Includes bibliographical references (v. 1, leaves 305-313).

Thesis (D.M.A.)--University of North Carolina at Greensboro, 2003. / Includes bibliographical references (leaves 137-140).

Thesis--University of Michigan. / Photocopy of typescript. Ann Arbor, Mich. : University Microfilms International, 1979. -- 21 cm. Includes bibliographical references (leaves 211-215).

Thesis (A. Mus. D.)--University of Illinois at Urbana-Champaign, 1985. / Typescript. Vita. Available on microfilm from University Microfilms. Includes bibliographical references (leaves 218-241).

Managing customer credit is an important issue for each commercial bank; therefore, banks take great care when dealing with customer loans to avoid any improper decisions that can lead to loss of opportunity or financial losses. The manual estimation of customer creditworthiness has become both time- and resource-consuming. Moreover, a manual approach is subjective (dependable on the bank employee who gives this estimation), which is why devising and implementing programming models that provide loan estimations is the only way of eradicating the ‘human factor’ in this problem. This model should give recommendations to the bank in terms of whether or not a loan should be given, or otherwise can give a probability in relation to whether the loan will be returned. Nowadays, a number of models have been designed, but there is no ideal classifier amongst these models since each gives some percentage of incorrect outputs; this is a critical consideration when each percent of incorrect answer can mean millions of dollars of losses for large banks. However, the LR remains the industry standard tool for credit-scoring models development. For this purpose, an investigation is carried out on the combination of the most efficient classifiers in credit-scoring scope in an attempt to produce a classifier that exceeds each of its classifiers or components. In this work, a fusion model referred to as ‘the Classifiers Consensus Approach’ is developed, which gives a lot better performance than each of single classifiers that constitute it. The difference of the consensus approach and the majority of other combiners lie in the fact that the consensus approach adopts the model of real expert group behaviour during the process of finding the consensus (aggregate) answer. The consensus model is compared not only with single classifiers, but also with traditional combiners and a quite complex combiner model known as the ‘Dynamic Ensemble Selection’ approach. As a pre-processing technique, step data-filtering (select training entries which fits input data well and remove outliers and noisy data) and feature selection (remove useless and statistically insignificant features which values are low correlated with real quality of loan) are used. These techniques are valuable in significantly improving the consensus approach results. Results clearly show that the consensus approach is statistically better (with 95% confidence value, according to Friedman test) than any other single classifier or combiner analysed; this means that for similar datasets, there is a 95% guarantee that the consensus approach will outperform all other classifiers. The consensus approach gives not only the best accuracy, but also better AUC value, Brier score and H-measure for almost all datasets investigated in this thesis. Moreover, it outperformed Logistic Regression. Thus, it has been proven that the use of the consensus approach for credit-scoring is justified and recommended in commercial banks. Along with the consensus approach, the dynamic ensemble selection approach is analysed, the results of which show that, under some conditions, the dynamic ensemble selection approach can rival the consensus approach. The good sides of dynamic ensemble selection approach include its stability and high accuracy on various datasets. The consensus approach, which is improved in this work, may be considered in banks that hold the same characteristics of the datasets used in this work, where utilisation could decrease the level of mistakenly rejected loans of solvent customers, and the level of mistakenly accepted loans that are never to be returned. Furthermore, the consensus approach is a notable step in the direction of building a universal classifier that can fit data with any structure. Another advantage of the consensus approach is its flexibility; therefore, even if the input data is changed due to various reasons, the consensus approach can be easily re-trained and used with the same performance.

abstract: The proper quantification and visualization of uncertainty requires a high level of domain knowledge. Despite this, few studies have collected and compared the roles, experiences and opinions of scientists in different types of uncertainty analysis. I address this gap by conducting two types of studies: 1) a domain characterization study with general questions for experts from various fields based on a recent literature review in ensemble analysis and visualization, and; 2) a long-term interview with domain experts focusing on specific problems and challenges in uncertainty analysis. From the domain characterization, I identified the most common metrics applied for uncertainty quantification and discussed the current visualization applications of these methods. Based on the interviews with domain experts, I characterized the background and intents of the experts when performing uncertainty analysis. This enables me to characterize domain needs that are currently underrepresented or unsupported in the literature. Finally, I developed a new framework for visualizing uncertainty in climate ensembles. / Dissertation/Thesis / Masters Thesis Computer Science 2016

Hailstones and Birdcages is a composition of approximately thirteen minutes' duration and is scored for two flutes and piccolo, two oboes and english horn, three Eb clarinets, E clarinete, bass clarinet, two bassoons, two Eb alto saxophones, Bb tenor saxophone, three Bb trumpets, four F horns, three trombones, euphonium, two tubas, and three percussionists. Four instruments--one each of flute, oboe, Bb clarinet, and trombone--are used in concertante like fashion, and there are prominent solo passages for the first bassoon, as well. The work is a single movement in three sections, fast - slow - fast, with ritornello. and employs a free use of the total chromatic. Technically, the work is within the capabilities of an above-average high school or average college wind ensemble

Cluster ensembles are collections of individual solutions to a given clustering problem which are useful or necessary to consider in a wide range of applications. The R package clue provides an extensible computational environment for creating and analyzing cluster ensembles, with basic data structures for representing partitions and hierarchies, and facilities for computing on these, including methods for measuring proximity and obtaining consensus and "secondary" clusterings. (author's abstract)

This study investigated relationships between 36 predictor variables and ensemble pitch and rhythm error discrimination ability. Precollege musical background and other demographic data were collected by means of the Musical Background Questionnaire. Musical achievement was measured by the Aliferis-Stecklein Music Achievement Test, College Midpoint Level. Undergraduate musical coursework data were obtained from transcripts. The criterion variables were measured by the Ramsey-Vincent Test of Instrumental Error Detection; a test of aural-visual pitch and rhythm error discrimination for full-score band music of medium difficulty. All three instruments were administered to 82 undergraduate music students. Subjects represented three Canadian universities and two community colleges. Pearson product-moment correlation tests were used to identify variables significantly related to musical ensemble error discrimination at the .10 level of significance. Eighteen variables were found to be significantly related to ensemble pitch error discrimination. Fourteen variables were found to be significantly related to ensemble rhythm error discrimination. Regression procedures were performed for each of the significant variables. These variables were then organized into blocks representing precollege musical background, other demographic variables, musical achievement, and undergraduate coursework. Regressions were performed for each of the blocks. Musical achievement, precollege musical background, demographic, and undergraduate coursework blocks of variables accounted for 5, 15, 35, and 21 percent of the variance in ensemble pitch error discrimination scores respectively. Musical achievement, precollege musical background, demographic, and undergraduate coursework blocks of variables accounted for 21, 16, 19, and 12 percent of the variance in ensemble rhythm error discrimination scores respectively. Combinations of variables from these blocks produced a linear model comprised of five demographic variables plus precollege choral experience that accounted for 42 percent of the variance in ensemble pitch error discrimination scores. Combinations of variables from the four blocks produced a linear model of ensemble rhythm error discrimination comprised of rhythmic discrimination, choice of a band instrument as one's major performance medium, composition as one's program major, and precollege band or orchestral experience. These four variables accounted for 32 percent of the variance in ensemble rhythm error discrimination scores. The variables selected for use in this study accounted for a substantial portion of the variance in error discrimination scores. To improve the predictive power of future studies, other variables need to be identified and included in the model. Ten conclusions were made regarding the prediction of ensemble error prediction ability. Three recommendations were made for improving error discrimination training and seven recommendations were made for future research in ensemble error discrimination. / Education, Faculty of / Graduate

L’objectif principal de cette thèse est d’étudier les structures périphériques des groupes relativement hyperboliques. En contraste avec l’hyperbolicité ordinaire, l’hyperbolicité relative est définie par rapport à une famille finie de sous-groupes, appelée structure périphérique. Dans cette thèse, on introduit et caractérise une classe de structures paraboliques étendues pour des groupes relativement hyperboliques. En particulier, on montre que si un groupe relativement hyperbolique agit de façon géométriquement finie sur son bord de Floyd, alors les structures paraboliques étendues se révèlent être les seules possibles.La thèse met également l’accent sur l’étude des sous-groupes relativement quasiconvexes, qui jouent un rôle important en théorie des groupes relativement hyperboliques. Grâce à la flexibilité des structures périphériques, la quasiconvexité relative d’un sous-groupe est caractérisée par rapport aux structures paraboliques étendues. En outre, les sous-groupes relativement quasiconvexes sont étudiés par des méthodes dynamiques en terme des groupes de convergence. Ceci nous conduit à obtenir un théorème décrivant l’intersection des ensembles limites pour une paire de sous-groupes relativement quasiconvexes ; et donner des preuves dynamiques de plusieurs résultats bien connus sur les sous-groupes relativement quasiconvexes. De plus, le nombre de classes de conjugaison de sous-groupes finis est étudié dans des groupes relativement hyperboliques. Dans le cas des groupes kleiniens, on obtient plusieurs résultats sur le lien entre les ensembles d’axes et la commensurabilité de deux groupes kleiniens. Un résultat de la thèse d’intérêt indépendant montre qu’un sous-groupe separable a la propriété d’empilement borné. Ceci implique que cette propriété est vraie pour tout sous-groupe d’un groupe polycyclique, répondant à une question de Hruska-Wise. / The main objective of this thesis is to study peripheral structures of relatively hyperbolic groups. In contrast with hyperbolicity, relative hyperbolicity is defined with respect to a finite collection of subgroups, which is referred to as a peripheral structure. In the thesis, we introduce and characterize a class of peripheralstructures: parabolically extended structures for relatively hyperbolic groups. In particular, it is shown that if a relatively hyperbolic group acts geometrically finitely on its Floyd boundary, then parabolically extended structures turn out to be the only possible ones. The thesis also focuses on the study of relatively quasiconvex subgroups, which play an important role in the theory of relatively hyperbolic groups. With the flexibility of peripheral structures, relative quasiconvexity of a subgroup is characterized with respect to parabolically extended structures. Moreover, relatively quasiconvex subgroups are studied using dynamical methods in terms of convergence group actions. This leads us to obtain a limit set intersection theorem for a pair of relatively quasiconvex subgroups, and give dynamical proofs of several well-known results on relatively quasiconvex subgroups. In addition, the number of conjugacy classes of finite subgroups is explored in relatively hyperbolic groups. In Kleinian groups, we prove several results on the relationship between the axes sets and commensurability of two Kleinian groups. A result of independent interest in the thesis is that a separable subgroup has the bounded packing property. This implies that the property is true for each subgroup of a polycyclic group, answering a question of Hruska-Wise.