Bin Lu. / "September 2004." / Thesis (Ph.D.)--Chinese University of Hong Kong, 2004. / Includes bibliographical references (p. 118-126) / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

Pan, Junhao. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2009. / Includes bibliographical references (leaves 121-135). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

Document clustering automatically organizes a document collection into distinct groups of similar documents on the basis of their contents. Most of existing document clustering techniques deal with monolingual documents (i.e., documents written in one language). However, with the trend of globalization and advances in Internet technology, an organization or individual often generates/acquires and subsequently archives documents in different languages, thus creating the need for multilingual document clustering (MLDC). Motivated by its significance and need, this study designs a Latent Semantic Indexing (LSI) based MLDC technique. Our empirical evaluation results show that the proposed LSI-based multilingual document clustering technique achieves satisfactory clustering effectiveness, measured by both cluster recall and cluster precision.

<p>Current frontiers in complex stochastic modeling of high-dimensional processes include major emphases on so-called functional data: problems in which the data are snapshots of curves and surfaces representing fundamentally important scientific quantities. This thesis explores new Bayesian methodologies for functional data analysis. </p><p>The first part of the thesis places emphasis on the role of factor models in functional data analysis. Data reduction becomes mandatory when dealing with such high-dimensional data, more so when data are available on a large number of individuals. In Chapter 2 we present a novel Bayesian framework which employs a latent factor construction to represent each variable by a low dimensional summary. Further, we explore the important issue of modeling and analyzing the relationship of functional data with other covariate and outcome variables simultaneously measured on the same subjects.</p><p>The second part of the thesis is concerned with the analysis of circadian data. The focus is on the identification of circadian genes that is, genes whose expression levels appear to be rhythmic through time with a period of approximately 24 hours. While addressing this goal, most of the current literature does not account for the potential dependence across genes. In Chapter 4, we propose a Bayesian approach which employs latent factors to accommodate dependence and verify patterns and relationships between genes, while representing the true gene expression trajectories in the Fourier domain allows for inference on period, phase, and amplitude of the signal.</p><p>The third part of the thesis is concerned with the statistical analysis of computer models (simulators). The heavy computational demand of these input-output maps calls for statistical techniques that quickly estimate the surface output at untried inputs given a few preliminary runs of the simulator at a set design points. In this regard, we propose a Bayesian methodology based on a non-stationary Gaussian process. Relying on a model-based assessment of uncertainty, we envision a sequential design technique which helps choosing input points where the simulator should be run to minimize the uncertainty in posterior surface estimation in an optimal way. The proposed non-stationary approach adapts well to output surfaces of unconstrained shape.</p> / Dissertation

Thesis (Ph. D.)--University of California, San Diego and San Diego State University, 2007. / Title from first page of PDF file (viewed May 29, 2007). Available via ProQuest Digital Dissertations. Vita. Includes bibliographical references (p. 99-105).

Thesis (Ph. D.)--University of Washington, 2006. / Vita. Includes bibliographical references (p. 115-121).

Le déterminisme de la formation de la grappe reste parmi les phénomènes les moins bien précisés de la physiologie de la Vigne. La question a été abordée de plusieurs façons mais les éléments de réponse obtenus ne permettent pas encore de se faire une idée synthétique des mécanismes fondamentaux qui produisent l’apparition des primordia inflorescentiels (PI) en année N-1 et qui pourraient régir jusqu’à 65 à 70% du rendement final en année N. Notre étude s’est donc focalisée sur (i) la caractérisation du processus de formation des PI dans les bourgeons latents de vigne (Vitis vinifera L. Merlot) et la redéfinition de la constitution du rendement, (ii) la compréhension des mécanismes d’initiation et de différenciation des PI et (iii) l’évaluation des impacts de différents facteurs (ABA exogène, lumière et effeuillage) sur ces mécanismes.Cette étude intégrale combinant des approches multi-échelles et multi-disciplinaires a permis dans un premier temps de mettre en évidence le déterminisme de la phase d’induction inflorescentielle dans la construction du rendement en identifiant trois phases clés du processus de formation des PI et en y intégrant une nouvelle notion qu’est le « degré de ramification des PI » ; puis dans un second temps, de proposer certains modes de régulation des mécanismes d’initiation et de différenciation des PI en identifiant trois types de marqueurs précoces de la fertilité potentielle : des marqueurs moléculaires (VvCLE25-1, VvTFL1a et VvMFT1), biochimiques (équilibre CKs/AIA, D-glucose et amidon), et environnementaux (ensoleillement et température journaliers cumulés). L’ensemble des résultats pourra contribuer au développement d’un outil prédictif précoce de la « fertilité potentielle » (et donc du rendement probable maximum) afin de mieux guider et gérer la taille et les pratiques culturales. / The grape cluster formation determinism remains the one of the undefined phenomena in grapevine physiology process. The topic has been tackled in multiple ways. However, based on all the known information, we still are not allowed to get a synthetic idea of the fundamental mechanisms that produce the appearance of inflorescential primordia in year N-1 and which could govern up to 65 to 70 % of final yield in year N. Our study focuses thus on (i) the characterization of IP formation process in grapevine latent buds (Vitis vinifera L. Merlot) and the redefinition of yield constitution, (ii) understanding IP initiation and differentiation mechanisms, and (iii) evaluating the impacts of different factors (exogenous ABA, light and leaf removal) on these mechanisms.This comprehensive study, which combines multi-scale and multi-disciplinary approaches, firstly allowed us to highlight the determinism of inflorescence induction phase in yield construction by identifying three key phases in IP formation process and by incorporating a new notion which is "IP branching degree". Secondly, it proposed certain modes of regulation of IP initiation and differentiation mechanisms by identifying three types of early markers of bud fruitfulness: molecular markers (VvCLE25-1, VvTFL1a and VvMFT1), biochemical markers (CKs/IAA balance, D-glucose and starch), and environmental markers (cumulative daily sunshine and temperature).The all-inclusive results contribute to the development of an early predictive tool of "bud fruitfulness" (and thereby the probable maximum yield) to better guide and manage the grapevine training and other cultural practices.

abstract: This dissertation examined Mexican American individuals' romantic relationships within two distinct developmental periods, adolescence and adulthood. Study 1 used latent class analysis to explore whether 12th grade Mexican Americans' (N = 218) romantic relationship characteristics, cultural values, and gender created unique romantic relationship profiles. Results suggested a three-class solution: higher quality, satisfactory quality, and lower quality romantic relationships. Subsequently, associations between profiles and adolescents' adjustment variables were examined via regression analyses. Adolescents with higher and satisfactory quality romantic relationships reported greater future family expectations, higher self-esteem, and fewer externalizing symptoms than adolescents with lower quality romantic relationships. Similarly, adolescents with higher quality romantic relationships reported greater academic self-efficacy and fewer sexual partners than adolescents with lower quality romantic relationships. Finally, adolescents with higher quality romantic relationships also reported greater future family expectations and higher academic self-efficacy than adolescents with satisfactory quality romantic relationships. To summarize, results suggested that adolescents engaged in three unique types of romantic relationships with higher quality being most optimal for their adjustment. Study 2 used latent growth modeling to examine marital partners' (N = 466) intra- and inter-individual changes of acculturative stress, depressive symptoms, and marital quality. On average across the seven years, husbands' acculturative stress remained steady, but wives' significantly decreased; partners' depressive symptoms remained relatively steady, but their marital quality significantly decreased. Although partners' experiences of acculturative stress were less similar than their experiences of depressive symptoms and marital quality, overall their experiences were interconnected. Significant spillover and crossover effects emerged between partners' initial levels of acculturative stress and depressive symptoms and between depressive symptoms and marital quality. Moreover, changes in husbands' depressive symptoms were negatively associated with changes in their marital quality. Overall, results suggested that partners' experiences were interconnected across time. / Dissertation/Thesis / Ph.D. Family and Human Development 2014

One of the more important social contexts in student life may be the student group. This study is based on a qualitative method to provide a deeper insight into how a student group describes the different phases it goes through and the functions of a student group. To answer the purpose, two research questions were asked: (1) How does a student group describe the different phases it goes through? (2) What manifest and latent functions can be distinguished in a student group? The empirical material for the study was gathered through five semi-structured interviews, in which the respondents were members of the same student group. I have analyzed the empirics by applying two theories, Merton's theory of latent and manifest functions and a theory of group processes by Lennéer Axelson and Thylefors. These two theories were chosen to answer the purpose and the research questions. Previous studies were also used to understand a student group's social significance and its manifest and latent functions. It appeared during the analysis of the material that the student group creates safety, belonging and social cohesion for the individual, the student group also created a study-based support. The student group that was studied also considered themselves as a group of friends. The specific purpose that existed at the beginning of the group constellation was changed when the group members created deeper relationships. The manifest function was that the respondents would start studying together, the manifest function developed latent functions. This study ends with a discussion on how a student group can be understood as an informal group and as an institutional group. This study also discuss what the university as an institution can do to support student groups. The discussion also presents what the university as an institution can do to support student groups. / En av studentlivets mer betydelsefulla sociala sammanhang kan vara studentgruppen. Denna studie utgår från en kvalitativ metod för att ge en djupare inblick i hur en studentgrupp beskriver de olika faser den går igen och en studentgrupps funktioner. För att besvara syftet ställdes två forskningsfrågor: (1) Hur beskriver en studentgrupp de olika faserna den går igenom? (2) Vilka manifesta och latenta funktioner kan urskiljas i en studentgrupp? Det empiriska materialet till studien samlades genom fem semistrukturerade intervjuer, där respondenterna var medlemmar i samma studentgrupp. Jag har analyserat materialet utifrån två teorier, Mertons teori om latenta och manifesta funktioner samt en teori om grupprocesser av Lennéer Axelson och Thylefors. Dessa två teorier valdes för att besvara syftet och forskningsfrågorna. Det framkom vid analysen av materialet att studentgruppen skapar trygghet, tillhörighet samt social sammanhållning för den enskilda individen, studentgruppen skapade även ett studiemässigt stöd. Studentgruppen som studerades betraktade även gruppen som ett kompisgäng och det konkreta syftet som fanns i början av gruppkonstellationen förändrades genom att gruppens medlemmar skapade djupare relationer. Den manifesta funktionen var att respondenterna skulle börja plugga tillsammans, den manifesta funktionen utvecklade latenta funktioner. Studien avslutas med en diskussion om hur en studentgrupp kan förstås, som en informell grupp och även en institutionell grupp. I diskussionen presenteras även vad universitetet som institution kan göra för att stödja studentgrupper.

Vibration-based condition monitoring is a key and crucial element for asset longevity and to avoid unexpected financial compromise. Currently, data-driven methodologies often require significant investments into data acquisition and a large amount of operational data for both healthy and unhealthy cases. The acquisition of unhealthy fault data is often financially infeasible and the result is that most methods detailed in literature are not suitable for critical industrial applications. In this work, unsupervised latent variable models negate the requirement for asset fault data. These models operate by learning the representation of healthy data and utilise health indicators to track deviance from this representation. A variety of latent variable models are compared, namely: Principal Component Analysis, Variational Auto-Encoders and Generative Adversarial Network-based methods. This research investigated the relationship between time-series data and latent variable model design under the sensible notion of data interpretation, the influence of model complexity on result performance on different datasets and shows that the latent manifold, when untangled and traversed in a sensible manner, is indicative of damage. Three latent health indicators are proposed in this work and utilised in conjunction with a proposed temporal preservation approach. The performance is compared over the different models. It was found that these latent health indicators can augment standard health indicators and benefit model performance. This allows one to compare the performance of different latent variable models, an approach that has not been realised in previous work as the interpretation of the latent manifold and the manifold response to anomalous instances had not been explored. If all aspects of a latent variable model are systematically investigated and compared, different models can be analysed on a consistent platform. In the model analysis step, a latent variable model is used to evaluate the available data such that the health indicators used to infer the health state of an asset, are available for analysis and comparison. The datasets investigated in this work consist of stationary and time-varying operating conditions. The objective was to determine whether deep learning is comparable or on par with state-of-the-art signal processing techniques. The results showed that damage is detectable in both the input space and the latent space and can be trended to identify clear condition deviance points. This highlights that both spaces are indicative of damage when analysed in a sensible manner. A key take away from this work is that for data that contains impulsive components that manifest naturally and not due to the presence of a fault, the anomaly detection procedure may be limited by inherent assumptions made in model formulations concerning Gaussianity. This work illustrates how the latent manifold is useful for the detection of anomalous instances, how one must consider a variety of latent-variable model types and how subtle changes to data processing can benefit model performance analysis substantially. For vibration-based condition monitoring, latent variable models offer significant improvements in fault diagnostics and reduce the requirement for expert knowledge. This can ultimately improve asset longevity and the investment required from businesses in asset maintenance. / Dissertation (MEng (Mechanical Engineering))--University of Pretoria, 2020. / Eskom Power Plant Engineering Institute (EPPEI) / UP Postgraduate Bursary / Mechanical and Aeronautical Engineering / MEng (Mechanical Engineering) / Unrestricted