With the continuous expansion of data availability in many large-scale, complex, and networked systems, such as surveillance, security, the Internet, and finance, it becomes critical to advance the fundamental understanding of knowledge discovery and analysis from raw data to support decision-making processes. Existing knowledge discovery and data analyzing techniques have shown great success in many real-world applications such as applying Automatic Target Recognition (ATR) methods to detect targets of interest in imagery, drug activity prediction, computer vision recognition, and so on. Among these techniques, Multiple-Instance (MI) learning is different from standard classification since it uses a set of bags containing many instances as input. The instances in each bag are not labeled | instead the bags themselves are labeled. In this area many researchers have accomplished a lot of work and made a lot of progress. However, there still exist some areas which are not covered. In this thesis, we focus on two topics of MI learning: (1) Investigating the relationship between MI learning and other multiple pattern learning methods, which include multi-view learning, data fusion method and multi-kernel SVM. (2) Dealing with the class imbalance problem of MI learning. In the first topic, three different learning frameworks will be presented for general MI learning. The first uses multiple view approaches to deal with MI problem, the second is a data fusion framework, and the third framework, which is an extension of the first framework, uses multiple-kernel SVM. Experimental results show that the approaches presented work well on solving MI problem. The second topic is concerned with the imbalanced MI problem. Here we investigate the performance of learning algorithms in the presence of underrepresented data and severe class distribution skews. For this problem, we propose three solution frameworks: a data re-sampling framework, a cost-sensitive boosting framework and an adaptive instance-weighted boosting SVM (with the name IB_SVM) for MI learning. Experimental results - on both benchmark datasets and application datasets - show that the proposed frameworks are proved to be effective solutions for the imbalanced problem of MI learning.

目標：本文旨在建構一種新型的模糊回歸模式，解決一类較複雜的模糊回歸問題。 研究方法:推廣局部加權回歸的思想，先從理論上構建新模型；然後借由模拟數據，從多個方面考察新模型的性質，并和其他模型做比較。 發現：局部加權回歸方法結合模糊隸屬度概念，使模糊回歸理論有更多的應用場合。 原創性：目前在模糊回歸領域的主流思想是通過線性規劃等方法來構建模型，而本文另闢蹊徑，首次從局部加權的角度構建了模糊回歸的新模型。 關鍵字: 模糊理論 模糊回歸分析 局部加權 / Objective: This paper aims to construct a new fuzzy regression model to solve a more complex fuzzy regression problem. Method: Build a new model by promoting the idea of locally weighted regression; Using simulated data to compare the new model with other models. Conclusion: The fuzzy membership degree concept combined with the locally weighted regression method makes the fuzzy regression theory have more applications. Originality: At present, the main idea in the field of fuzzy regression is to construct models by means of linear programming. In this paper, a new model of fuzzy regression is constructed from the perspective of locally weighted method for the first time. Keyword: Fuzzy theory、 Fuzzy regression、Locally weighted method

La réflexion sur l‟utilisation de la politique budgétaire comme outil de stabilisation et de relance connaît un net regain d‟intérêt ces dernières années. Après près de trois décennies qui ont vu la dominance des idées néo-Classique, la récente crise financière des années 2008 a consacré le retour aux idées keynésiennes sur l‟efficacité de l‟outil budgétaire. Cette thèse s‟intéresse à ce thème et essaie de caractériser la politique budgétaire dans le contexte des pays en développement et son objectif final est de préciser dans quelle mesure cet outil de politique économique serait efficace pour ces pays. Le chapitre 2 traite de la question des effets des politiques budgétaires surprises. Autrement dit, et à partir d‟une modélisation en VAR structurels, cette partie se pose la question de savoir si le budget peut être utilisé de façon surprise pour relancer une économie et quels sont les défis que pose une telle mesure dans le contexte d‟une économie en développement. Le troisième chapitre à partir d‟un modèle de gravité analyse les relations entre la situation budgétaire dans les économies avancées ainsi que celle des pays émergents et les flux d‟investissement vers les économies à revenu intermédiaire. Cette étude montre qu‟un effet d‟éviction entre pays (développés et émergents) existe mais aussi que l‟économie mondiale tend vers un nouveau paradigme. Le dernier chapitre quant à lui étudie la cyclicité des politiques budgétaires pour un échantillon de pays d‟Afrique subsaharienne et d‟Amérique latine. La méthode choisie a permis de suivre l‟évolution de la procyclicité des politiques budgétaires d‟année en année et de montrer que les pays en développement surtout africains progressivement adoptent des politiques de plus en plus disciplinées et prudentes / The use of fiscal policy as a stabilization and stimulus tool face a renewed interest from analyst and policy makers. After almost three decades where neo-Classical ideas were dominant, the recent financial crisis (late 2007) marked the reborn of Keynesian ideas on the importance of the State budget during economic downturns. This dissertation focuses on this issue and provides with stylized facts of fiscal policies in developing economies, and the main aim being to be able to say whether fiscal policy is an efficient political economy tool. Chapter 2 focuses on the issue of unanticipated fiscal measures on the economy. Using a structural VAR approach it investigates whether unanticipated budget measures can be used to stimulate a declining economy and what kind of challenges and threats this strategy imposes to public authorities. Chapter 3, relying on a gravity model, analyses the relationship between emerging and advanced economies fiscal aggregates and capital flows. It shows that there exists a “global” crowding out effect of investment towards emerging markets and, most important is that world economy is entering into a new paradigm. The last chapter from a panel of Sub-Saharan African and Latin American economies studies the issue of fiscal procyclicality. The empirical strategy has allowed us on a yearly basis to characterise the cyclical behaviour of fiscal policies in both set of countries. It has been shown that developing countries especially African ones are adopting progressively more prudent and disciplined policies.

The hypothesis claims that full service carriers (FSCs) will be forced to change their operating models. Additionally, the trends which form aviation market of the future are explored. The investigation is based on a financial assessment of ten important airlines incorporated in four different regions across the globe, between 2005 and 2014. According to the findings the trend of liberalization will continue, thus FSCs will continue losing market share to low cost carriers (LCCs) and as such, they will be made to change the way they operate in order to survive. The future aviation market is defined by a reshuffle of demand towards emerging countries, further rationalization of operating models and consolidation.

The aim of this paper is to investigate various versions of the Phillips curve and their inflation forecasting ability for Euro Area. We consider autoregressive distributed lag models and use two types of trend estimation -- successive (the trend is estimated before the remaining parameters are) and join, using exponential smoothing. The versions of the Phillips curve are evaluated by rolling and recursive window methods, various selection criteria for lag variables and different combination of the inflation indicators. To evaluate the forecasted values, we calculate the RMSE in three 7-year periods: 1993-1999 (run up Euro area), 2000-2006 (stable inflation period) and 2007-2013 (financial crisis). According to all our modifications, we find some models which achieve satisfying results in terms of the RMSE, albeit not for all forecasting periods. We notice that some models are satisfactory only in the stable period however not in the periods with low inflation and vice versa.

Diploma thesis deals with the valuation of the automotive company Automotive lighting, s.r.o for the management needs. The estimated value refers to October 2012. The basis for the valuation is strategic and financial analysis. The results will be taken into account for establishment of the financial plan, which will be used in the valuation process. Emphasis is placed on yield methods using free cash flow to the firm(FCFF) and economic value added EVA. Both methods are compared with book value of the company.

Today's dynamic period requires sufficient amount of high-quality information to make right decision. Valuation of a company may be one of the sources. It is a result of complex valuation of a company and its activity from different points of view. Information gained during the process valuation can be used to effective management of a company, too. The main aim of this thesis is to evaluate the company's environment, the market where the business runs as well as the analyses its future possibilities. Following historic data, strategic and financial analyses will be carried out and these results will serve for company value determination, using the yield method. Gained information will be valuable as far as company's management as well as investors are concerned. The result lies in historical progression, contemporary situation, potential and future prediction valuation. Recognized company's value will serve as the final output.

The dissertation thesis aims to establish a complex theoretical basis for a measurement of customer's profitability which can be easily used in the practice. The main reason for that is the absence of such publication in the current academic sphere. A combination of management accounting and knowledge of banking enables to achieve the aim. The thesis itself is closely linked to the banking practice. There are derived customer's profitability indicators as modifications of the popular RAROC in which a customer margin is used instead of net profit. There is designed a simplified way of operational cost allocation. Questionnaires and interviews with senior representatives of seven Czech banks helped to identify the most significant labor-intensity factors of corporate customers. The description of principal features of risk costs and derivation of formulas for profitability variances are also the part of the thesis. The theoretical part is followed by the practical one where a contemporary banking practice of corporate customer's profitability measurement on the Czech market is explored. It identified three weak points -- cost allocation, profitability variances and one-year horizon of a calculation. At the end, the theoretical basis is applied on an existing customer portfolio and the result is compared with a currently used customer's profitability measurement.

Scandinavian Terrain Vehicles AB (STV) konstruerar och bygger bandvagnar för civilt bruk. I utbudet ingår även vad STV kallar Mid Life Improvement (MLI) vilket syftar i att renovera Hägglunds BV206. En del av MLI-programmet är byte av motor och växellåda där de gamla komponenterna från sent 80-tal till tidigt 90-tal byts ut mot ny motor och växellåda. Den nya kombinationen av motor och växellåda möter militär standard och blir därför en onödigt kostsam del då många av STV:s kunder skulle vara nöjda även med ett enklare alternativ. För att minska kostnaden för MLI-programmet och därmed potentiellt utöka kundkretsen undersöktes möjligheterna till att hitta nya leverantörer av motor och växellåda. Som alternativ undersöktes möjligheterna till drift genom diesel- eller Ottomotor samt transmission av typerna automatisk växellåda, hydrostatisk och elektrisk transmission. Sökning av komponenter utgick från att undersöka vilka komponenttillverkare som används av liknande applikationer. Genom en kriterieviktsmetod blev det rekommenderade lösningsalternativet motorn Cummins ISF 2.8 120 kW och växellådan Allison 1000SP. Kombinationen levererar som mest 1384 Nm och vid 5200 rpm 185 Nm till fördelningsväxeln. Resultatet styrks av att motorn enligt tillverkaren är lämpad för fordon såsom mindre lastbilar och terrängfordon samt att växellådor av samma tillverkare används till bland annat BAE Systems Hägglunds bandvagn BvS10. / Scandinavian Terrain Vehicles AB (STV) is a company that designs and builds tracked vehicles for civilian usage. One of their services is what they call Mid Life Improvement (MLI) which serves to recondition Hägglunds BV206’s. A significant part of the MLI-program is the exchange of engine and transmission where components from the late 80’s to early 90’s are replaced with new parts. The new parts consist of a combination of engine and gearbox that meets military standards and is therefore an expensive part especially since many of STV’s customers would settle with a more elementary alternative. To reduce costs within the MLI-program and potentially expand clientele STV sought to find new suppliers for engine and gearbox. Considered alternatives for replacement were Otto- and Diesel engines in coupled with a transmission in the form of either an automatic gearbox, a hydrostatic or an electric transmission. By examining which suppliers were being used by similar applications and selection through a weighted criteria method, the engine Cummins ISF 2.8 120 kW and transmission Allison 1000SP were found to be the best suited alternative. The combination delivers to the distribution gearbox a maximum torque of 1384 Nm and 185 Nm at 5200 rpm. The result is attested by the fact that the engine is according to the manufacturer suitable for applications like trucks and utility vehicles together with that the same brand of transmission is used by BAE Systems Hägglunds in their BvS10.

Integrating decision analytical techniques in geographic information systems (GIS) can help remove the two primary obstacles in spatial decision making: inaccessibility to required geographic data and difficulties in synthesizing various criteria. I developed a GIS model to assist people seeking optimal residential locations. Fuzzy set theory was used to codify criteria for each factor used in evaluating residential locations, and weighted linear combination (WLC) was employed to simulate users' preferences in decision making. Three examples were used to demonstrate the applications in the study area. The results from the examples were analyzed. The model and the ArcGIS Extension can be used in other geographic areas for residential location selection, or in other applications of spatial decision making.