The creation and development ofcustomer relationship has become a strong marketing tool for most suppliers inthe current business world. In addition, the development of relationships hasbecome the main topic for market researchers since they believe that, customersand suppliers are valuable assets of a firm. Most suppliers also face thedilemma of how they should adapt to their customers needs and requirementsbecause adaptations always tighten the bonds of both the supplier and thecustomer and make them more dependent on each other. In order to identify thereasons why firms place much importance on the development of customerrelationships, the authors have undertaken a qualitative study. This study wasbased on why ABB Automation develops relationship with its customer’s base. Thedata collected was analyzed to identify the reasons behind the company’sdecisions.

本研究主要的目的是在探討不為社會一般人所知道與關心的HIV感染者，在感染之後適應疾病的過程，面臨的困難，需要的服務，以及生活適應的問題。進而提出社會工作專業幫助感染者度過生命黑暗期的可能策略。只有當感染者不再因為感染而感受到生死的糾葛，不再因為感染，而被社會道德的節勒得喘不過氣來，防制愛滋的蔓延，才有可能。 本研究總結對HIV感染者的觀察與深入訪談後發現，感染者在知道自己感染之後，最能夠在它們身上感受到的是對曝光的擔心；孤獨的感受；死亡的陰影；以及階段任務受阻等心理層面上的壓力與現實上生活適應的困難交織而成的生死結。 感染者普遍說來，都十分擔心曝光。而孤獨的感受對感染者來說，一是因為擔心曝光、自殘形穢的自我封閉而造成；一是因為感到社會的遺棄，缺乏規屬感所產生。而由於感染的必死性，讓感染者不論目前的身體狀況如何，都生活在死亡的陰影之下，而這樣的死亡陰影，亦會影響到感染者求生與求死的意念。階段性任務受阻的感受，人生的任務無法達成，則是感染者普遍的遺憾。 而在心理層面的擔憂之外，感染者難以避免的會遭遇到實際生活適應上的困難。一般說來感染者會產生生活水準降低的問題；醫療相關的問題；擔心發病；空間剝奪的問題；因擔心曝光或告知所產生的心理壓力問題。 但是生死結並非在所有感染者身上都能起相同的作用，換句話說，不同的感染者對生死結的感受，會隨著他們的感染途徑；自覺社會烙印；原本的社會地位；感染事實被知悉的範圍以及得知感染的時間等前置因素而有所不同。這些前置因素，因為影響到感染者對生死結的感受，進而也會影響到感染者所採取的因應策略。感染者所採取的因應策略大致可以區分為幾種：改變生活方式與態度；告知；死亡的準備；特殊化以及利他。 針對以上感染者的生活適應困境本研究提出以下幾點建議，提供相關單位與實務工作者參考。給相關單位建議的部分有：1、基本生活水準的維持；2、健全醫療服務網；3、專業藥師提供用藥諮詢；4、在相關醫療院所聘用專任涉工員；5、義工的訓練與督導；6、設立協談專線；7、跨部會的定期會議。而對提供社會工作直接服務者的建議是：1、非批判式的接納；2、避免特殊化；3、多傾聽。

Doctor of Philosophy / Department of Computing and Information Sciences / Doina Caragea / The large volume of data generated in the recent years has created opportunities for discoveries in various fields. In biology, next generation sequencing technologies determine faster and cheaper the exact order of nucleotides present within a DNA or RNA fragment. This large volume of data requires the use of automated tools to extract information and generate knowledge. Machine learning classification algorithms provide an automated means to annotate data but require some of these data to be manually labeled by human experts, a process that is costly and time consuming. An alternative to labeling data is to use existing labeled data from a related domain, the source domain, if any such data is available, to train a classifier for the domain of interest, the target domain. However, the classification accuracy usually decreases for the domain of interest as the distance between the source and target domains increases. Another alternative is to label some data and complement it with abundant unlabeled data from the same domain, and train a semi-supervised classifier, although the unlabeled data can mislead such classifier. In this work another alternative is considered, domain adaptation, in which the goal is to train an accurate classifier for a domain with limited labeled data and abundant unlabeled data, the target domain, by leveraging labeled data from a related domain, the source domain. Several domain adaptation classifiers are proposed, derived from a supervised discriminative classifier (logistic regression) or a supervised generative classifier (naïve Bayes), and some of the factors that influence their accuracy are studied: features, data used from the source domain, how to incorporate the unlabeled data, and how to combine all available data. The proposed approaches were evaluated on two biological problems -- protein localization and ab initio splice site prediction. The former is motivated by the fact that predicting where a protein is localized provides an indication for its function, whereas the latter is an essential step in gene prediction.

Agriculture is one of the most important human activities providing food and more agricultural goods for seven billion people around the world and is of special importance in sub-Saharan Africa. The majority of people depends on the agricultural sector for their livelihoods and will suffer from negative climate change impacts on agriculture until the middle and end of the 21st century, even more if weak governments, economic crises or violent conflicts endanger the countries’ food security. The impact of temperature increases and changing precipitation patterns on agricultural vegetation motivated this thesis in the first place. Analyzing the potentials of reducing negative climate change impacts by adapting crop management to changing climate is a second objective of the thesis. As a precondition for simulating climate change impacts on agricultural crops with a global crop model first the timing of sowing in the tropics was improved and validated as this is an important factor determining the length and timing of the crops´ development phases, the occurrence of water stress and final crop yield. Crop yields are projected to decline in most regions which is evident from the results of this thesis, but the uncertainties that exist in climate projections and in the efficiency of adaptation options because of political, economical or institutional obstacles have to be considered. The effect of temperature increases and changing precipitation patterns on crop yields can be analyzed separately and varies in space across the continent. Southern Africa is clearly the region most susceptible to climate change, especially to precipitation changes. The Sahel north of 13° N and parts of Eastern Africa with short growing seasons below 120 days and limited wet season precipitation of less than 500 mm are also vulnerable to precipitation changes while in most other part of East and Central Africa, in contrast, the effect of temperature increase on crops overbalances the precipitation effect and is most pronounced in a band stretching from Angola to Ethiopia in the 2060s. The results of this thesis confirm the findings from previous studies on the magnitude of climate change impact on crops in sub-Saharan Africa but beyond that helps to understand the drivers of these changes and the potential of certain management strategies for adaptation in more detail. Crop yield changes depend on the initial growing conditions, on the magnitude of climate change, and on the crop, cropping system and adaptive capacity of African farmers which is only now evident from this comprehensive study for sub-Saharan Africa. Furthermore this study improves the representation of tropical cropping systems in a global crop model and considers the major food crops cultivated in sub-Saharan Africa and climate change impacts throughout the continent. / Landwirtschaft ist eine der wichtigsten menschlichen Aktivitäten, sie stellt Nahrungsmittel und andere landwirtschaftliche Produkte für weltweit 7 Milliarden Menschen zur Verfügung und ist in den Ländern Afrikas südlich der Sahara von besonderer Bedeutung. Die Mehrheit der afrikanischen Bevölkerung bestreitet ihren Lebensunterhalt in der Landwirtschaft und wird von Klimaänderungen stark betroffen sein. Die Doktorarbeit ist durch die Frage motiviert, wie sich von Klimamodellen vorhergesagte Temperaturerhöhungen und sich verändernde Niederschlagsverteilungen auf die landwirtschaftliche Vegetation auswirken werden. Die Forschungsfragen in diesem Kontext beschäftigen sich mit regionalen Unterschieden von Klimaänderungen und ihren Auswirkungen auf die Landwirtschaft und mit möglichen Anpassungsstrategien die mit geringem technischem Aufwand genutzt werden können. In diesem Zusammenhang wird schnell deutlich, dass Daten über die komplexen landwirtschaftlichen Systeme in Afrika südlich der Sahara häufig nur selten vorhanden sind, aus fragwürdigen Quellen stammen oder von schlechter Qualität sind. Die Methoden und Modelle zur Untersuchung der Auswirkungen von Klimaänderungen auf die Landwirtschaft werden zudem ausschließlich in Europa oder Nordamerika entwickelt and häufig in den temperierten Breiten aber seltener in tropischen Gebieten angewendet. Vor allem werden globale, dynamische Vegetationsmodelle in Kombination mit Klimamodellen eingesetzt um Änderungen in der landwirtschaftlichen Produktion auf Grund von Klimaänderungen in der zweiten Hälfte des 21.Jahrhunderts abzuschätzen. Die Ergebnisse der Arbeit zeigen einen mittleren Ertragsrückgang für die wichtigsten landwirtschaftlichen Pflanzen um 6% bis 24% bis 2090 je nach Region, Klimamodell und Anpassungsstrategie. Dieses Ergebnis macht deutlich, dass Landwirte die negativen Folgen von Klimaänderungen abschwächen können, wenn sie die Wahl der Feldfrucht, die Wahl des Anbausystems und den Aussaattermin an geänderte Klimabedingungen anpassen. Die Arbeit stellt methodische Ansätze zur Berechung des Aussaattermins in temperierten und tropischen Gebieten (Kapitel 2) sowie zur Simulation von Mehrfachanbausystemen in den Tropen vor (Kapitel 3). Dabei werden wichtige Parameter für das globale, dynamische Vegetationsmodell LPJmL überprüft und neu berechnet. Es zeigt sich, dass das südliche Afrika und die Sahelregion die am stärksten betroffenen Regionen sind, vor allem aufgrund von Niederschlagsänderungen, weniger aufgrund von Temperaturerhöhungen. In den meisten anderen Teilen, vor allem Zentral- und Ostafrikas bedingen Temperaturerhöhungen Rückgänge der Erträge (Kapitel 4). Diese Arbeit leistet einen wichtigen und umfassenden Beitrag zum Verständnis der Auswirkung von Klimaänderung auf die landwirtschaftliche Vegetation und damit zu einem großen Teil auf die Lebensgrundlage von afrikanischen Landwirten.

Lorsqu’un sujet est soumis à des perturbations à la marche au sol ou sur tapis roulant, les réactions posturales et les ajustements posturaux anticipatoires s’améliorent progressivement avec la répétition des perturbations, entraînant une diminution de la perte d’équilibre. L’objectif du mémoire était d’analyser les ajustements posturaux et de l’équilibre anticipatoire entre des perturbations variées et aux déclenchements imprédictibles avec ou sans habituation au préalable. Deux études ont été effectuées pour le présent mémoire dans lesquelles 24 sujets jeunes en santé (moyenne ± écart type : 25,3 ±2,3 ans) ont été inclus. Les participants marchaient sur un tapis roulant à double courroie à marche confortable, sous différentes conditions : sans perturbation (Etudes 1 et 2), avec perturbations répétées (Etude 2), avec perturbations aléatoires (Etudes 1et 2). Les perturbations étaient générées par des accélérations ou des décélérations de différentes amplitudes, d’une des courroies du tapis pendant la phase d’appui droite ou gauche de façon aléatoire tous les 8 à 20 pas. Les données cinématiques et cinétiques ont été enregistrées. La difficulté à maintenir l’équilibre augmentait à la fin des essais de perturbations aléatoires comparativement au début avec (Etude 2) et sans habituation préalable (Etude 1) et à la fin des deux derniers essais d’habituations aux perturbations de type décélération (Etude 2). Aucun changement de l’équilibre n’a été observé lors des essais d’habituation aux perturbations de type accélération. L’augmentation de la difficulté à maintenir l’équilibre était due principalement à une diminution de la distance entre le centre de pression et la base de support. Aucun changement de l’équilibre n’a été observé lors des essais d’habituation aux perturbations de type accélération. Les changements observés lors des essais aléatoires seraient une stratégie posturale non-spécifique adoptée par les participants afin de répondre aux perturbations et en particulier lorsqu’elles sont imprévisibles. / When repeated postural perturbations are induced while walking on the floor or on a treadmill, reactive and proactive postural adjustments improve progressively. The objective of the study was to analyse proactive postural and balance adjustments between varied perturbations with unpredictable onsets with and without a previous period of habituation. Two studies have been conducted in which 24 young healthy subjects (mean ± standard deviation: 25.3 ± 2.3 years) participated. Participants walked on a slip-belt treadmill at comfortable speed under different conditions: without perturbations (studies 1 and 2), with repeated perturbations (study2), and with random perturbations (studies 1 and 2). Perturbations consisted in an accelerations or a decelerations of various magnitude, of one of the belt of the treadmill during right or left single support stance every 8 to 20 steps. Three dimensional whole-body kinematic and kinetic data were recorded. The difficulty to maintain balance increased at the end of trials with random perturbations compared to the beginning with (study 2) and without (study 1) habituation to the perturbations. Balance difficulty also increased at the end of the trials with repeated deceleration perturbations (Study 2). No change were observed during repeated acceleration perturbation trials. The increase of the difficulty to maintain balance was mainly due to a decrease between the center of pressure and the base of support. No change were observed during repeated acceleration perturbation trials. Changes observed during random trials may be a non-specific strategy adopted by the participants to respond to perturbations, particularly when they were unpredictable.