Made available in DSpace on 2019-03-29T23:32:55Z (GMT). No. of bitstreams: 0 Previous issue date: 2012-08-30 / The aim of this study is to establish a parallel between four bioinspired metaheuristic algorithms (namely, Genetic Algorithm, Differential Evolution, Particle Swarm Optimization, and Harmony Search), testing them with continuous optimization benchmark functions and assessing them based on data collected about the quality of the solutions they generate and based on the way they explore the search space. Novel techniques are proposed to collect data about the modus operandi of the algorithms using Voronoi Diagrams and through a division of the search space in bi-dimensional regions using Self-Organizing Maps. The use of these maps as a Visual Data Mining tool aims to process the resulting data and identify the available clusters. We wish to understand the influence of parameter calibration on the search behavior of the metaheuristic algorithms, as well as their sensitivity to dimension changes of the optimization functions they are solving. Furthermore, behavior profiles can be established and patterns can be defined, allowing us to analyze in detail the similarities/dissimilarities between the algorithms. All metaheuristics were implemented step-by-step to allow us to understand their inner workings, peculiarities, and also facilitate the tracking of their search behavior along their execution on the optimization problems. Keywords: Bioinspired Metaheuristics. Information Visualization. Visual Data Mining. Self-Organizing Maps. Kohonen Maps. Voronoi Diagrams. / Este trabalho tem por finalidade traçar um paralelo entre quatro algoritmos bio-inspirados de cunho meta-heurístico (a saber: Algoritmo Genético, Evolução Diferencial, Otimização por Enxame de Partículas e Busca Harmônica), testando-os em funções de benchmark de otimização contínua e avaliando-os com base em dados coletados sobre a qualidade das soluções geradas e sobre o modo de exploração do espaço de busca. São propostas aqui técnicas de coleta de informações sobre o modus operandi dos algoritmos que envolvem Diagramas de Voronoi e divisão do espaço de busca em regiões bidimensionais utilizando Mapas Auto-organizáveis. O uso desses mapas como ferramenta de Mineração Visual de Dados tem como objetivo avaliar os dados gerados e identificar os agrupamentos que foram formados. Procuramos entender a influência da alteração dos valores dos parâmetros das meta-heurísticas no seu comportamento ao longo do tempo, bem como a sua sensibilidade às mudanças de dimensão das funções de otimização que estão solucionando. Dessa forma, almeja-se traçar perfis de comportamento e definir a posteriori um padrão para possibilitar a análise mais detalhada das similaridades/dissimilaridades entre as abordagens. Todos os algoritmos utilizados foram implementados passo-a-passo a fim de entender as suas particularidades, esclarecer melhor seus mecanismos internos e facilitar o rastreamento do comportamento dos mesmos no ato da resolução dos problemas. Palavras-chave: Meta-heurísticas Bioinspiradas. Visualização da Informação. Mineração Visual de Dados. Mapas Auto-Organizáveis. Mapas de Kohonen. Diagramas de Voronoi.

This thesis explores the relation between the normative structures brought to bear on the evaluation of life and the way in which the coming-into-being of living organisms is fundamentally understood. It provides a new analysis and critique of the standard concept of ‘flourishing’ in neo-Aristotelian meta-ethics, by uncovering the underlying interpretation of organismic becoming on which it relies, and showing how the turn to a ‘constructivist’ conception of development in contemporary biological theory both disrupts this underlying metaphysics, and provides resources for re-thinking flourishing on a fundamentally different basis. The central claim is that we should turn from a view in which life is given a form to fulfil, and becoming is the process of its fulfilment, to one in which living is the process of creating a way in the world, as life goes along.

Cheung Shu Fai. / "August 2000." / Thesis (Ph.D.)--Chinese University of Hong Kong, 2000. / Includes bibliographical references (p. 117-123). / 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.

Yau Man Yin. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (leaves 44-57). / Abstract in English and Chinese. / Chapter CHAPTER 1. --- INTRODUCTION --- p.1 / Review on Stress and Coping --- p.1 / Coping --- p.2 / Research on Stress and Coping --- p.3 / Using SMIs to Study Stress and Coping --- p.5 / Occupational Stress --- p.6 / Table 1 --- p.9 / Reasons for Meta-Analysis --- p.10 / Content As a Moderator --- p.11 / Identification of Stressors as a Moderator --- p.12 / Massed Vs Distributed Session as a Moderator --- p.13 / Time Interval as Moderator --- p.15 / Outcomes --- p.16 / Chapter CHAPTER 2. --- METHOD --- p.18 / Literature Search --- p.18 / Inclusion Criteria --- p.18 / Meta-Analysis Procedures --- p.19 / Coding --- p.21 / Publication Bias --- p.21 / Chapter CHAPTER 3. --- RESULTS --- p.22 / Overall --- p.22 / Table 2 --- p.23 / Moderators --- p.24 / Table 3 --- p.24 / Table 4 --- p.26 / Table 5 --- p.26 / Table 6 --- p.27 / Table7 --- p.29 / Table 8 --- p.30 / Table 9 --- p.31 / Chapter CHAPTER 4: --- DISCUSSION --- p.32 / Workshop Content --- p.34 / Identification of Stress --- p.35 / Massed and Distributed Practice --- p.36 / Time interval --- p.37 / Implications --- p.38 / Limitations and Future Directions --- p.40 / Concluding Remarks --- p.42 / REFERENCES --- p.44

Au Ka Yan. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (leaves 35-47). / Abstract in English and Chinese. / ABSTRACT --- p.1 / 摘要 --- p.2 / ACKNOWLEDGEMENT --- p.3 / Chapter CHAPTER 1. --- INTRODUCTION --- p.7 / Leadership Competencies --- p.7 / Importance of Leadership Training --- p.8 / Use of Meta-Analysis --- p.9 / Meta-Analysis on Leadership Training --- p.9 / Applying Leadership Theories to Leadership Training --- p.10 / Spacing Effect --- p.12 / Methodological Rigor of Program Design --- p.13 / Hypotheses --- p.16 / Chapter CHAPTER 2. --- METHOD --- p.18 / Literature Search --- p.18 / Inclusion Criteria --- p.18 / Coding Procedure --- p.19 / Meta-Analysis Procedure --- p.20 / Chapter CHAPTER 3. --- RESULTS --- p.22 / Overall Effectiveness --- p.22 / Moderators --- p.24 / Chapter CHAPTER 4. --- DISCUSSION --- p.27 / Discussion on Key Findings --- p.27 / Leadership Training Is Effective in General --- p.27 / Applying Theories Can Enhance the Effectiveness --- p.28 / Spaced Distribution Outperforms Mass Distribution --- p.29 / Characteristics of Measurement May Distort the Effect Size --- p.29 / Theoretical and Practical Implications --- p.31 / Implications for Researchers --- p.31 / Implications for Practitioners --- p.32 / Limitations and Further Studies --- p.33 / What Is Leadership Training Actually? --- p.33 / Meta-Analysis Is Far From Perfect --- p.34 / REFERENCES --- p.35

Cheung Wai-leung. / "July 2002." / Thesis (Ph.D.)--Chinese University of Hong Kong, 2002. / Includes bibliographical references (p. 110-129). / 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.

Background There now exist a range of efficacious options for the treatment of psychosis in mental healthcare. The importance of recovery, empowerment, dignity and choice among patients with severe mental health diagnoses are important topics in contemporary research and practice. This thesis presents a meta-analytic review followed by a randomised controlled experimental study. These address distinct but related questions which aim to further our understanding of the choices available for intervention in psychosis and whether intervention may improve the ability of psychosis patients to make those choices. Aims The first objective aimed to offer a comprehensive review of the effectiveness of social skills training (SST), which is a psychological intervention for psychosis. SST has fallen out of favour in the UK and is not widely implemented in practice. We hypothesised that SST would demonstrate superiority for the negative symptoms of psychosis. The second objective was to determine whether decision-making capacity regarding treatment among psychosis patients could be improved by the application of a brief psycho-educational intervention targeting the jumping-to-conclusions (JTC) bias, which is a commonly observed cognitive bias in psychosis. We hypothesised that the intervention would improve decision-making capacity. Methods Firstly, a series of 70 meta-analyses are presented in a systematic review assessing the efficacy of social skills training across a number of psychosis outcome domains: positive symptoms, negative symptoms, general symptoms, overall symptoms and social functioning outcomes. Secondly, a randomised controlled experimental study is presented in which 36 psychosis patients in NHS Lanarkshire and 1 in NHS Dumfries & Galloway were allocated to receive either a brief meta-cognitive training (MCT) intervention or an non-specific control presentation lecture. Capacity was assessed at baseline and post-treatment while the impact of the intervention upon capacity was estimated by ANCOVA. Mediation analyses assessed whether changes in the JTC bias mediated outcome. Findings In the meta-analytic review, SST demonstrated superiority over treatment as usual (TAU. g=0.3), active controls (g=0.2-0.3) and comparators pooled (g=0.2- 0.3) for negative symptoms; and over TAU (g=0.4) and comparators pooled (g=0.3) for general psychopathology. In the randomised controlled study, MCT demonstrated large effects on two capacity outcomes; overall capacity (d=0.96, p < .05) and appreciation (d=0.87, p < .05). Exploratory analyses suggested a mediating effect of JTC (d=0.64, p < .05). Interpretation SST demonstrates a magnitude of effect for negative symptoms similar to those commonly reported for CBT for positive symptoms and may have potential for wider implementation in mental healthcare settings. The randomised controlled study suggests that psycho-educational interventions targeting capacity have clinical utility and may be developed for implementation. Limitations included lack of blinding, no fidelity checks and inclusion based on clinical diagnosis therefore a larger randomised controlled trial addressing these limitations is warranted.

Duplication of the genome and successful mitotic cell division requires the coordinated activity of hundreds of proteins. Many are known, but a complete list of the components of the cell cycle machinery is still lacking. This thesis describes a series of data driven analyses to assemble a comprehensive list of genes induced during the human cell cycle. To start with, a meta-analysis of previous transcriptomics studies revealed a larger number of cell cycle genes consistently expressed across multiple human cell types than previously reported. Following this observation, the cell cycle transcriptome was further investigated with the generation of a new time-course microarray dataset on normal human dermal fibroblasts (NHDF) undergoing synchronised cell division. Network cluster analysis of these data identified transcripts whose expression was associated with different stages of cell cycle progression. Co-expression of these transcripts was then analysed using a complementary dataset that included genome-wide promoter expression of a wide range of human primary cells. This resulted in the identification of a core set of 545 cell cycle genes, mainly associated with G1/S to M phases, which showed a high degree of co-expression across all cell types. Expression of 75% of these genes was also found conserved in mouse, as revealed by the analysis of a new microarray experiment generated from mouse fibroblasts. Gene Ontology and motif enrichment analysis validated the list with significant enrichments for terms and transcription factor biding sites linked with cell cycle biology. Toward a better interpretation of these 545 genes, a meticulous manual annotation exercise was carried out. Unsurprisingly, the majority of these genes were known to be involved in S and M phases-associated processes, however 50 genes were functionally uncharacterised. A subset of 36 of these were then taken forward for subcellular localisation assays. These studies were performed by transfection of human embryonic kidney cells (HEK293T) with GFP-tagged cDNA clones leading to the finding of four uncharacterised proteins co-localising with the centrosome, a crucial organelle for normal cell cycle progression. This thesis represents an attempt in documenting the genes actively transcribed and therefore likely involved in the processes associated with cell cycle, hence providing a comprehensive catalogue of its key components. In so doing, I have also identified a significant number of new genes likely to contribute to this central process vital in health and disease.

abstract: The occurrence of micro-and nanoplastic (MNP) debris in the environment is a research area of considerable public health concern. Various combinations of methods for extraction, isolation, and quantification of MNP have been applied but literature studies evaluating the appropriateness and efficacy of these protocols are lacking. A meta-analysis of the literature (n=134; years 2010-2017) was conducted to inventory and assess the appropriateness of methodologies employed. Some 30.6% of studies employed visual identification only, which carried a calculated misidentification error of 25.8-74.2%. An additional 6.7% of studies reported counts for particles smaller than the cutoff value of the selected collection pore size, and 9.7% of studies utilized extraction solution densities which exclude some of the polymers commonly occurring in the environments investigated. A composite value of data vulnerability of 43.3% was determined for the sample, indicating considerable weaknesses in the robustness of information available on MNP occurrence and type. Additionally, the oxidizing solutions documented in the literature frequently were deemed unsuccessful in removing interfering organic matter. Whereas nanoplastics measuring <1 µm in diameter are likely principal drivers of health risk, polymer fragments reported on in the literature are much larger, measuring 10+ µm in diameter due to lack of standardized methods. Thus, current inventories of MNP in the environmental MNP feature data quality concerns that should be addressed moving forward by using more robust and standardized techniques for sampling, processing and polymer identification to improve data quality and avoid the risk of misclassification. / Dissertation/Thesis / Masters Thesis Civil, Environmental and Sustainable Engineering 2018

O problema do caixeiro viajante (PCV) é um problema clássico de otimização que possui diversas variações, aplicações e instâncias. Encontrar a solução ótima para muitas instâncias desse problema é geralmente muito difícil devido o alto custo computacional. Vários métodos de otimização, conhecidos como meta-heurísticas (MHs), são capazes de encontrar boas soluções para o PCV. Muitos algoritmos baseados em diversas MHs têm sido propostos e investigados para diferentes variações do PCV. Como não existe um algoritmo universal que encontre a melhor solução para todas as instâncias de um problema, diferentes MHs podem prover a melhor solução para diferentes instâncias do PCV. Desse modo, a seleção a priori da MH que produza a melhor solução para uma dada instância é uma tarefa difícil. A pesquisa desenvolvida nesta tese investiga o uso de abordagens de meta-aprendizado para selecionar as MHs mais promissoras para novas instâncias de PCV. Essas abordagens induzem meta-modelos preditivos a partir do treinamento das técnicas de aprendizado de máquina em um conjunto de meta-dados. Cada meta-exemplo, em nosso conjunto de meta-dados, representa uma instância de PCV descrita por características (meta-atributos) do PCV e pelo desempenho das MHs (meta-atributo alvo) para essa instância. Os meta-modelos induzidos são usados para indicar os valores do meta-atributo alvo para novas instâncias do PCV. Vários experimentos foram realizados durante a investigação desta pesquisa e resultados importantes foram obtidos / The traveling salesman problem (TSP) is a classical optimization problem that has several variations, applications and instances. To find the optimal solution for many instances of this problem is usually a very hard task due to high computational cost. Various optimization methods, known as metaheuristics (MHs), are capable to generate good solutions for the TSP. Many algorithms based on different MHs have been proposed and investigated for different variations of the TSP. Different MHs can provide the best optimization solution for different TSP instances, since there is no a universal algorithm able to find the best solution for all instances. Thus, a priori selection of the MH that produces the best solution for a given instance is a hard task. The research developed in this thesis investigates the use of meta-learning approaches to select the most promising MHs for new TSP instances. These approaches induce predictive meta-models from the training of machine learning techniques on a set of meta-data. In our meta-data, each meta-example is a TSP instance described by problem characteristics (meta-features) and performance of MHs (target meta-features) for this instance. The induced meta-models are used to indicate the values of the target meta-feature for new TSP instances. During the investigation of this research, several experiments were performed and important results were obtained