published_or_final_version / Education / Master / Master of Education

許慶豫. / 論文(哲學博士)--香港中文大學, 2001. / 參考文獻 (p. 346-364) / 中英文摘要. / Available also through the Internet via Dissertations & theses @ Chinese University of Hong Kong. / 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. / Xu Qingyu. / Lun wen (Zhe xue bo shi)--Xianggang Zhong wen da xue, 2001. / Can kao wen xian (p. 346-364) / Zhong Ying wen zhai yao.

Business process automation improves organizations’ efficiency to perform work. In existing business process management systems, process instances run independently from each other. However, synchronizing instances carrying similar characteristics, i.e., sharing the same data, can reduce process execution costs. For example, if an online retailer receives two orders from one customer, there is a chance that they can be packed and shipped together to save shipment costs. In this paper, we use concepts from the database domain and introduce data views to business processes to identify instances which can be synchronized. Based on data views, we introduce the concept of batch regions for a context-aware instance synchronization over a set of connected activities. We also evaluate the concepts introduced in this paper with a case study comparing costs for normal and batch processing. / Die Automatisierung von Geschäftsprozessen unterstützt Unternehmen, die Ausführung ihrer Prozesse effizienter zu gestalten. In existierenden Business Process Management Systemen, werden die Instanzen eines Prozesses völlig unabhängig voneinander ausgeführt. Jedoch kann das Synchronisieren von Instanzen mit ähnlichen Charakteristiken wie z.B. den gleichen Daten zu reduzierten Ausführungskosten führen. Zum Beispiel, wenn ein Onlinehändler zwei Bestellungen vom selben Kunden mit der gleichen Lieferanschrift erhält, können diese zusammen verpackt und versendet werden, um Versandkosten zu sparen. In diesem Papier verwenden wir Konzepte aus dem Datenbankbereich und führen Datensichten für Geschäftsprozesse ein, um Instanzen zu identifizieren, welche synchronisiert werden können. Auf Grundlage der Datensichten führen wir das Konzept der Batch-Regionen ein. Eine Batch-Region ermöglicht eine kontext-bewusste Instanzen-Synchronisierung über mehrere verbundene Aktivitäten. Das eingeführte Konzept wird mit einer Fallstudie evaluiert, bei der ein Kostenvergleich zwischen der normalen Prozessausführung und der Batchverarbeitung durchgeführt wird.

The purpose of this study is to determine what differences, if any, exist in reading, math and spelling achievement scores as measured by the Stanford Achievement Tests of the second grade students who were grouped homogeneously (i.e., were assumed to have similar academic abilities) and the second grade students who were heterogeneously grouped (i.e., were grouped according to no particular criteria) when compared to their first grade scores. These two groups of second graders, each consisting of four classes with approximately twenty-five students in each class, provide an excellent opportunity for research study for the purpose stated above because they were all exposed to the same four teachers, similar methods and materials. All of these students involved in this experiment attended the same school in the first grade and most of them attended the same school in kindergarten.

The thesis contains a description and a legal analysis of the European Economic Interest Grouping. The goal is achieved through the method of description, analysis and comparison. The first chapters of the thesis are devoted to incorporation of the Grouping, development of legislation relating to the Grouping, applicable system of law and the topic of legal nature of the Goruping. The way of how to incorporate a Grouping is described in the following chapter. Furthermore, the rights and obligations deriving from the membership in a Grouping are explained. The final chapters are devoted to the process of dissolution of the Grouping and to popularity of the Grouping throughout the EU countries.

A segmentação de imagens e, em visão computacional, uma tarefa de grande importância, para a qual existem várias abordagem. A complexidade de tais abordagens está relacionada à natureza da imagem e também ao grau de precisão da segmentação, que e um conceito bastante subjetivo, normalmente associado a semelhança que apresenta a segmentaçã produzida pela visão humana. Na segmentação de imagens baseada em algoritmos de agrupamento em grafos, geralmente os pixels da imagem compôem os nós do grafo e as arestas representam a similaridade entre estes nós. Assim, a segmentação pode ser obtida por meio do agrupamento dos nós do grafo. É importante salientar, no entanto, que as técnicas de agrupamento em grafos surgiram no contexto de reconhecimento de padrões, cujo objetivo primario era o tratamento de dados diversos que não envolviam imagens. O uso de tais tecnicas para a segmentação de imagens e relativamente recente e revela alguns problemas desaadores. O primeiro deles é a deficiente escalabilidade de alguns métodos, o que impede o seu uso efetivo em imagens de altas dimensões. Outra questão é a falta de estudos que avaliam as medidas de similaridade na montagem do grafo e critérios que aferem a qualidade do agrupamento para a área específica de segmentação de imagens. Em outras palavras, faltam na literatura análises mais específicas que indiquem quais algoritmos de agrupamento em grafos são mais efetivos para a segmentação de imagens e que procurem associar (ou correlacionar) as várias medidas de similaridade e métricas de qualidade de agrupamento que produzam segmentações mais precisas. Neste trabalho é apresentada a avaliação de 6 algoritmos de agrupamento em grafos formulados em base a 3 categorias identificadas (agrupamento espectral, algoritmos de particionamento multinível e algoritmos para detectar comunidades) e aplicadas na segmentação automática de imagens de cenas naturais com grandes dimensões. Esta avaliação objetiva aferir, sobretudo, a qualidade da segmentação, a escalabilidade, o desempenho de 7 funções de similaridade formuladas, e também visa corroborar a existência da correlação entre a qualidade do agrupamento e a qualidade da segmentação. Para reduzir o esforço computacional e contribuir com a escalabilidade dos algoritmos formulados é utilizado um algoritmo de pré-processamento (SLIC) que agrupa váarios pixels da imagem em uma unica região (superpixels), o que contribui para reduzir o tamanho do grafo e, consequentemente, reduzindo o custo computacional do agrupamento. Os resultados demostram que os algoritmos formulados LP (Label Propagation) e FG (Fast Greedy) apresentam boa escalabilidade e boa qualidade de segmentação. Seis das sete funções de similaridade avaliadas apresentam um bom desempenho, independentemente do algoritmo de agrupamento empregado. É mostrado também que exites correlação entre a medida de qualidade de agrupamento conhecido como índice de silhueta e a qualidade de segmentação, ou seja, quanto maior o valor de silhueta, melhor a segmentação. A qualidade de segmentação foi avaliada quantitativamente, utilizando-se um conjunto de imagens segmentadas manualmente / Image segmentation is an important task within computer vision for which many approaches are available. The complexity of such approaches are intrinsically related with the nature of the image and also the desired accuracy aimed at. Image segmentation accuracy, however, is a subjective concept and is normally associated with how much it resembles segmentation produced by the human vision system. In graphbased clustering image segmentation algorithms, pixels are normally represented as nodes and edges convey the similarity between such nodes. Hence, segmentation may be attained by means of grouping node of a graph. It is important, though, to point out that graph-based clustering techniques rst appeared in the context of pattern recognition and its primary data source were not images. The usage of such techniques for image segmentation is a recent trend and poses some challenge issues. The first is the poor scalability that many methods exhibit, impairing its application in images of larger dimensions. Another issues is that lack of studies that assess the goodness of similarity measures employed in graph computation and also clustering quality criteria assessments for the specic area of image processing. In other words, there is no evidences in the literature on how effective graph-based clustering algorithms are for image segmentation and no studies that associate similarity functions and clustering quality metrics with image processing quality. This work presents an evaluation of six graph-based clustering algorithms according to three major categories found in the literature (spectral clustering, multi layer partitioning algorithms and community detectors) applied to automatic segmentation of image of larger dimensions. This evaluation takes into account segmentation quality, scalability, the performance of seven similarity functions and, nally, bring some insights on the correlation between clustering and segmentation quality. To reduce computational costs and enhance scalability of the methods we employ a pre processing algorithm (SLIC) which combines many pixels into one single region (superpixel). This contributes to reduce the graph size and, consequently, the cost of clustering. Results have shown that the LP (Label Propagation) and FG (Fast Greedy) algorithms exhibit good scalability and good segmentation. Six out of the seven similarity functions presented good performance, regardless the clustering algorithm. We also have shown that there is correlation between clustering quality and image segmentation quality, when the Silhouette measure is employed. That is, the higher the Silhouette values, the better the segmentation. Image segmentation quality has been quantitatively assessed by means of ground-truth data or user segmented images

Submitted by (ana.araujo@ufrpe.br) on 2017-02-16T16:09:53Z No. of bitstreams: 1 Gheysa Coelho Silva.pdf: 1993608 bytes, checksum: fb8764341d9f83a94b65cdbfa671783a (MD5) / Made available in DSpace on 2017-02-16T16:09:53Z (GMT). No. of bitstreams: 1 Gheysa Coelho Silva.pdf: 1993608 bytes, checksum: fb8764341d9f83a94b65cdbfa671783a (MD5) Previous issue date: 2008-07-14 / Conselho Nacional de Pesquisa e Desenvolvimento Científico e Tecnológico - CNPq / The work had as aimed (1): to quantify the genetic divergence between seven varieties patterns and eleven clones sugar-cane RB through techniques united and multivariate with base in ten characters agroindustrial.; (2) to evaluate the relationships among the agronomic and technological characters about the productivity in cane and sugar, and to identify your direct and indirect effects, to aid the process of selection of clones RB of sugar-cane. The experiment was developed in the agricultural area in sugar mill Trapiche - Sirinhaém-PE, during the agricultural year 2005/2006. The experimental design was randomized blocks with four repetitions and eighteen genotypes constituted by seven varieties patterns. Were appraised in cane-plant for the agricultural production the characters NC, DC, AC, PMC and TCH. To determinate of the technological characteristics BC, PC, FIB, PUR and TPH were measured. The widespread distance of Mahalanobis was used as dissimilarity measure. Were considered the correlations, and they were made the trail analyses. Were applied the hierarchical method of medium connections (UPGMA) and the method of optimization of Tocher. The results evidenced that the methodology allows to identify genotypes of larger genetic divergence for use in programs of improvement of the sugarcane, allowing to suggest possible crossings between the most divergent clones and productive RB992542 and RB992558 with the variety SP79-1911 and of the clone RB992558 with the variety RB92579. The characters PC, TCH, BC and AC were the main ones decisive in the quantification of the genetic divergence. The characters PMC, AC and DC should be considered in the selection of clones RB seeking improvement in the sugarcane productivity, as well as TCH and PC they can be selected seeking increase of the productivity of sugar. Indirectly, the selection through height and diameter of stems makes possible efficient answer for improvement of the sugarcane productivity. / Objetivou-se com este trabalho: (1) quantificar a divergência genética entre sete variedades-padrões e onze clones RB de cana-de-açúcar, por meio de técnicas uni e multivariadas, com base em dez caracteres agroindustriais. (2) avaliar as relações entre os caracteres agronômicos e tecnológicos sobre a produtividade em cana e açúcar e identificar seus efeitos diretos e indiretos para auxiliar o processo de seleção de clones RB de cana-de-açúcar. O experimento foi conduzido na área agrícola da Usina Trapiche - Sirinhaém/PE, durante o ano agrícola 2005/2006. O delineamento experimental utilizado foi o de blocos casualizados com quatro repetições e dezoito genótipos. Foram avaliados em cana-planta para a produção agrícola os caracteres NC, DC, AC, PMC e TCH. Para determinação das características tecnológicas foram mensuradas BC, PC, FIB, PUR e TPH. A distância generalizada de Mahalanobis foi utilizada como medida de dissimilaridade. Foram aplicados o método hierárquico de ligações médias (UPGMA) e o método de otimização de Tocher. Estimaram-se as correlações e fizeram-se as análises de trilha. Os resultados evidenciaram que a metodologia permite identificar genótipos demaior divergência genética para utilização em programas de melhoramento da cana-de- açúcar, permitindo sugerir possíveis cruzamentos entre os clones mais divergentes e produtivos RB992542 e RB992558 com a variedade SP79-1911 e do clone RB992558 com a variedade RB92579. Os caracteres PC, TCH, BC e AC foram os principais determinantes na quantificação da divergência genética. Os caracteres PMC, AC e DC devem ser considerados na seleção de clones RB visando melhoria na produtividade de cana-de-açúcar, assim como TCH e PC podem ser selecionados visando aumento da produtividade de açúcar.Indiretamente, a seleção via altura e diâmetro de colmos possibilita resposta eficiente para melhoria da produtividade de cana-de-açúcar.

The purpose of this study was to determine if there was a significant difference in total scores on the Mathematical Self-Efficacy Scale, the mathematics task self-efficacy portion of the scale, and the math-related school subjects self-efficacy portion of the scale for middle school students between students assigned to a homogeneously grouped accelerated math class and students assigned to a heterogeneously grouped math class. The instrument used to gather information for thus study on student self-efficacy was the Mathematics Self-Efficacy Scale (MSES). The MSES measures 2 domains of mathematics-related behaviors and capabilities. The Mathematics Task Self-Efficacy scale is designed to measure the level of confidence the student would have when successfully completing the given task. The Math-Related School Subjects Self-Efficacy scale is designed to measure the level of confidence the student would have when successfully completing a college level course with a final grade of an A or B. The 2 parts of the MSES may be individually scored or holistically scored to obtain a total score representing overall mathematical self-efficacy. Descriptive and inferential statistics were used to analyze the data for the 9 research questions. Participants in the study were randomly assigned to the heterogeneous or homogeneous groups by their schools and were not controlled by the researcher. Students within the groups were chosen as participants based on their math ability and scores on the seventh grade TCAP test. At the time of the survey these students attended either a K-8 elementary school or a middle school in Northeast Tennessee. The population consisted of 357 gifted or talented eighth grade math students in 6 school districts in Northeast Tennessee. The results of this study does not support or discourage the practice of acceleration by retaining 7 of the 9 null hypotheses that there are no significant difference in self-efficacy scores between homogeneous grouped eighth grade math students who were placed in accelerated coursework by taking Algebra I and those students who were heterogeneously grouped in a regular eighth grade math class.

The ability to count people from video is a challenging problem. The scientific challenge arises from the fact that although the task is pretty well defined, the imaging scenario is not well constrained. The background scene is uncontrolled. Lighting is complex and varying. And, image resolution, both in terms of spatial and temporal is usually poor, especially in pre-stored surveillance videos. Passive counting of people from video has many practical applications such as in monitoring the number of people sitting in front of a TV set, counting people in an elevator, counting people passing through a security door, and counting people in a mall. This has led to some research in automated counting of people. The context of most of the work in people counting is in counting pedestrians in outdoor settings or moving subjects in indoor settings. There is little work done in counting of people who are not moving around and very little work done in people counting that can handle harsh variations in illumination conditions. In this thesis, we explore a design that handles such issues at pixel level using photometry based normalization and at feature level by exploiting spatiotemporal coherence that is present in the change seen in the video. We have worked on home and laboratory dataset. The home dataset has subjects watching television and the laboratory dataset has subjects working. The design of the people counter is based on video data that is temporally sparsely sampled at 15 seconds of time difference between consecutive frames. Specific computer vision methods used involves image intensity normalization, frame to frame differencing, motion accumulation using autoregressive model and grouping in spatio-temporal volume. The experimental results show: The algorithm is less susceptible to lighting changes. Given an empty scene with just lighting change it usually produces a count of zero. It can count in varying illumination conditions. It can count people even if they are partially visible. Counts are generated for any moving objects in the scene. It does not yet try to distinguish between humans and non-humans. Counting errors are concentrated around frames with large motion events, such as a person moving out from a scene.

Le présent travail de thèse a pour but d’étudier et d’améliorer la logistique du groupement hospitalier de Mulhouse sud-Alsace (GHRMSA). Notre étude se focalise sur les problématiques liées à la pharmacie à usage interne de l’hôpital Émile Muller de Mulhouse, car cette dernière est destinée à devenir la pharmacie centrale du groupement. Cette nouvelle considération implique une centralisation de certains processus pharmaceutiques tels que la gestion et la stérilisation des dispositifs médicaux nécessaires au bon déroulement des interventions chirurgicales pratiquées quotidiennement dans les différents centres de soins du groupement.Ce travail est donc divisé en deux parties distinctes ayant pour dénominateur commun la gestion des dispositifs médicaux.En effet, dans un premier temps nous envisageons l’optimisation des méthodes d'implantation pour le stockage de ces dispositifs. Les particularités du milieu hospitalier par rapport au milieu industriel nous permettent de mettre en place des méthodes d’agencement de stock dans le but de minimiser la distance moyenne parcourue par les agents lors de la phase de collecte. Tout d'abord, nous proposons un modèle mathématique linéaire, puis nous présentons une heuristique dédiée, mais aussi des méthodes hybridées combinant cette heuristique avec un algorithme génétique puis un branch and bound. Afin d’affiner ce travail, et de se projeter dans l’avenir, nous proposons ensuite une méthode d'optimisation de la collecte d’une liste de dotations au sein du stock prenant en compte des contraintes liées à l’utilisation d’un véhicule de transport et d’un stockage multi niveau, à savoir : l’impossibilité de faire des demi-tours, le respect du sens de circulation et le respect de la charge maximale transportable. Chacune de ces contraintes peut être prise en compte ou ignorée pour représenter au mieux la réalité.Dans un second temps, nous proposons des méthodes d’ordonnancement des interventions chirurgicales pratiquées par le GHRMSA. La particularité de ce travail et de se focaliser d’avantage sur l’assignation des ressources, c’est à dire des boites de dispositifs médicaux. Le but est de renforcer les liens décisionnels existants entre la pharmacie et les différents blocs opératoires de chaque site afin de lisser la charge de travail du service de stérilisation et d’assurer une meilleure réactivité en cas d’urgences ou d’imprévus. Pour résoudre ce problème d’ordonnancement avec affectation de ressources, nous définissons un modèle mathématique linéaire et proposons un algorithme d’optimisation par essaim particulaire pour minimiser le nombre de boites de dispositifs médicaux nécessaires. Enfin, nous étendons le problème initial avec la prise en compte du transport inter-sites. Nous détaillons alors une méthode pour définir des plannings d’interventions et de livraison des dispositifs médicaux pour minimiser le coût global en fonction des caractéristiques du système de transport. / The purpose of this thesis is to study and improve the logistical aspects of the hospitals group named "Groupement Hospitalier de Mulhouse Sud-Alsace" (GHRMSA). Our study focuses on the problematics of the pharmacy of the Emile Muller hospital at Mulhouse in France, which is dedicated to become the central pharmacy of the group. This new consideration implies to centralize several pharmaceutical processes such as the medical devices management which are mandatory for surgical operations. This work is divided into two different parts on the management of surgical devices.Indeed, as a first step we study the optimization of the layout to store surgical devices. The special characteristics of the pharmaceutical environment compared to the industrial one are used to propose methods to design layouts of the warehouse in order to minimize the average distance traveled by the employees during the picking phase. First, we define a linear mathematical model and a dedicated heuristic but also two other methods hybridizing the latter with genetic algorithms and branch and bounds. In order to improve this work and prepare the future, we also propose methods to optimise the routing of the employees to completes endowment lists using a transport vehicle with a multi level storing policy. The use of such a vehicle implies the following constraints: u-turn impossibility, imposed traffic direction and respect of the transport capacity. Any of theses constraints can be considered or not in order to create a model as near as possible of the real world.As a second step, we propose surgical operations scheduling methods. This work focuses on the assignment of the needed surgical devices boxes. The purpose is to improve the decision between the operating theaters of each site of the hospital group and the sterilization service in order to smooth the sterilization workload and to improve the reactivity in case of emergencies or any other unexpected problems. To solve the scheduling problem, we define a linear mathematical model as well as a particle swarm optimization algorithm to create surgical operations plannings which minimize the number of needed surgical devices boxes. Moreover, we extend the initial problem with the transport between hospital locations, and present a method to create plannings for surgical operations and for the delivery of the surgical boxes to minimize the global cost considering the parameters of the transport system.