《水滸傳》的創作從南宋初年至明代中葉，這四百多年來流傳的水滸故事，經過說書人的講演鋪陳、文人的潤飾增色，逐漸成熟完整，匯聚成一部精彩絕倫的通俗小說。小說設計了一百零八位英雄，這些好漢被賦予了不同出身、獨特相貌，以及各式穿著，繪製出一幅多彩多姿的水滸圖卷。《水滸傳》在人物登場時，往往以一套贊賦介紹人物從頭到腳的衣著，為讀者在腦海中畫出一個具體而鮮明的形象。由於「服飾」是一個人的「社會肌膚」，透過「服飾」我們可得知其身分地位、審美觀念、性格投射與所處的社會風尚，甚至能反映群體文化。《水滸傳》服飾琳瑯滿目，筆者透過文本的分析，將一百零八好漢的衣著，分為首服、上衣、下裳與足服、腰佩四類，逐一考掘形制內容、演化，宋、元、明三代服制與人物的穿用取向。其次探討服飾的質料與顏色，將質料分為棉麻、絲織、毛皮、金屬四類，研究服飾表現的物質文化；服色則採紅、青、白、黑、黃五色分述，闡發《水滸傳》服飾的穿搭美學。把人物服飾對應《水滸傳》的創作背景，瞭解小說反映出來宋、元、明三代紡織工藝的發達，帶動商品經濟活絡，形成豐富多樣的衣飾時尚以及崇奢僭禮的風氣。最後將服飾與人物形象結合探討，論述小說如何運用服飾勾勒《水滸傳》臉譜，研究服飾對人物外在的身分界定，與文本內在的書寫意涵。本論文從服飾的角度審視《水滸傳》，從中歸納出服飾是人物刻劃的途徑，是小說渲染情節的方法，也是社會風俗的具體反映，更是梁山好漢犯禁思想的表現，總結出《水滸傳》是通俗小說服飾書寫的轉型階段。 / The creation of The Water Margin went through a roughly 400-year span, starting from the beginning of Southern Song to the mid-Ming Dynasty. Stories of The Water Margin, performed, polished, refined and compiled by storytellers and poets, were made into one epic popular fiction, within which 108 outlaws, endowed with distinct backgrounds, appearances, and dress, made an animated landscape of Water Margin. The Water Margin introduces most of its characters’ complete outfits through a set of words of praise on their debuts, thereby portraying solid, bright images in its readers’ mind. Dress are socialized skin from which we tell one’s social status, aesthetics, projected nature, contemporary trends, and even collective culture from those of others’. The Water Margin presents 108 outlaws in a wide array of clothing which, in my textual analyses, was categorized into caps, garments, skirts and footwear, and waist ornaments. I explore the contents of their designs and transformations over time, and how people in Song, Yuan and Ming Dynasties designed and decided their dress in terms of wearing and functioning. Next, I’d discuss the textile—categorized into cotton and gunny, silk, hide and leather, and metal, in order to study the material culture embedded in dress—along with colors—red, cyan, white, black, and yellow included, so as to elaborate on the aesthetics of dressing. Comparing dress with the social context of The Water Margin’s creation gives insight to the highly-developed textile crafting in Song, Yuan, and Ming Dynasty, a crafting which promoted a booming commodity economy and thereby formed a miscellaneous dressing fashion and a luxury-worshiping, etiquette-violating atmosphere. Combining dress with the characters, I reason how each figure was shaped with what he or she wore, and study how one’s dressing defined his/her extrinsic identity and the textual implications inherited in the fiction. This thesis aims at scrutinizing The Water Margin through dress, concluding that dressing is a path to characters’ depictions, a manner with which the plot was embellished, a concrete reflection on social customs, as well as a representation of the 108 outlaws’ thinking of breaking prohibitions. The Water Margin, therefore, was at the stage of dressing depiction’s transformation in popular fictions.

Maize (Zea mays L.) is the most important grain crop in the country. Approximately 12 million tons of maize grain is produced annually on approximately 2.5 million ha of land. However, increased farming intensity can lead to fragmentation of habitat and has a tendency to decrease the biodiversity of an area. Therefore, to ensure the continued functionality of agro-ecosystems, methods in agriculture must be assessed and adapted when necessary to ensure the persistence of biological diversity. Unfortunately, the effect of crop production on species diversity and composition in South Africa is still relatively unknown, and no baseline data exists with which to gauge the possibility of unknown extinction risks of important biological elements. The objectives of this study were to compare plant and arthropod diversity patterns and species turnover of maize agro-ecosystems between biomes (grassland and savanna) and along a maize field-field margin gradient (MAFFMAG). Surveys of maize agro-ecosystems were conducted in six provinces of South Africa, namely North-West, Mpumalanga, KwaZulu-Natal, Limpopo, Free State and the Eastern Cape. Repeated measures ANOVA revealed a significantly lower plant and arthropod species diversity and richness in maize fields compared to field margins. Non-metric multidimensional scaling revealed that arthropod species composition differed between biomes although not along MAFFMAGs, indicating that arthropod species composition is dependent on biome rather than distance from maize field. Floristic data revealed unique species compositions for maize fields and field margins and also for biomes. Furthermore, maize fields and field margins of grassland sites were more similar in plant species composition than the savanna localities, suggesting higher regional beta diversity for savanna regions. Spearman‘s rank order correlations revealed generally positive but weak or no relationships between plant and arthropod diversity. This study provides baseline data for identification, monitoring and conservation of priority species and will allow the future evaluation of ecosystem services provided by plants and associated arthropods, especially natural enemies of pests, in maize agro-ecosystems. / MSc (Environmental Sciences), North-West University, Potchefstroom Campus, 2014

This thesis examines three aspects of the commentary edition of the Chinese vernacular novel Shuihu Zhuan written by Ming Dynasty literatus Jin Shengtan (ca. 1610-1661), analyzing three of the most innovative features that the commentary brings to our understanding of the novel, and what Jin Shengtan desired for the reader of his commentary to understand. The first chapter looks at a series of techniques that Jin outlines in the preliminary "How to Read" section of the commentary (dufa), where the techniques are shown to be very similar in focus and style to the literary theory of narratology as written about by Gerard Genette through a sample comparison of five of the techniques with varying characteristics of narratology. The second chapter looks at how Jin Shengtan constructs the image of the author, Shi Nai'an, through both his interlineal commentary (jiapi) and his preliminary chapter commentary (zongpi). We see through this analysis that Jin Shengtan has gone against the tradition of shu er bu zuo -- a Confucian tradition that relegates the position of the author to the background of his work -- and has brought the author into a position of prominence through his construction of the image of an unparalleled genius. The third and final chapter looks at the idea of "heroism" (xia) and how Jin's commentary reworks the way many of the primary characters of the novel and their heroic actions are seen and interpreted, focusing especially on the characters of Wu Song, Lu Zhishen, Song Jiang and Li Kui, where we see that Jin's commentary focuses on parallels between the heroes such as Wu Song and Lu Zhishen in the first portion of the novel, while switching to a more juxtapositional perspective in the latter half of the novel through Song Jiang and Li Kui. / Graduate / 0305 / 0332 / mblsm00@gmail.com

We consider the problem of learning the parameters of a structured output prediction model, that is, learning to predict elements of a complex interdependent output space that correspond to a given input. Unlike many of the existing approaches, we focus on the weakly supervised setting, where most (or all) of the training samples have only been partially annotated. Given such a weakly supervised dataset, our goal is to estimate accurate parameters of the model by minimizing the regularized empirical risk, where the risk is measured by a user-specified loss function. This task has previously been addressed by the well-known latent support vector machine (latent SVM) framework. We argue that, while latent SVM offers a computational efficient solution to loss-based weakly supervised learning, it suffers from the following three drawbacks: (i) the optimization problem corresponding to latent SVM is a difference-of-convex program, which is non-convex, and hence susceptible to bad local minimum solutions; (ii) the prediction rule of latent SVM only relies on the most likely value of the latent variables, and not the uncertainty in the latent variable values; and (iii) the loss function used to measure the risk is restricted to be independent of true (unknown) value of the latent variables. We address the the aforementioned drawbacks using three novel contributions. First, inspired by human learning, we design an automatic self-paced learning algorithm for latent SVM, which builds on the intuition that the learner should be presented in the training samples in a meaningful order that facilitates learning: starting frome easy samples and gradually moving to harder samples. Our algorithm simultaneously selects the easy samples and updates the parameters at each iteration by solving a biconvex optimization problem. Second, we propose a new family of LVMs called max-margin min-entropy (M3E) models, which includes latent SVM as a special case. Given an input, an M3E model predicts the output with the smallest corresponding Renyi entropy of generalized distribution, which relies not only on the probability of the output but also the uncertainty of the latent variable values. Third, we propose a novel learning framework for learning with general loss functions that may depend on the latent variables. Specifically, our framework simultaneously estimates two distributions: (i) a conditional distribution to model the uncertainty of the latent variables for a given input-output pair; and (ii) a delta distribution to predict the output and the latent variables for a given input. During learning, we encourage agreement between the two distributions by minimizing a loss-based dissimilarity coefficient. We demonstrate the efficacy of our contributions on standard machine learning applications using publicly available datasets.

This thesis addresses the challenges associated with thermal management systems (TMS) evaluation and selection in the conceptual design of hypersonic, air-breathing vehicles with sustained cruise. The proposed methodology identifies analysis tools and techniques which allow the proper investigation of the design space for various thermal management technologies. The design space exploration environment and alternative multi-objective decision making technique defined as Pareto-based Joint Probability Decision Making (PJPDM) is based on the approximation of 3-D Pareto frontiers and probabilistic technology effectiveness maps. These are generated through the evaluation of a Pareto Fitness function and Monte Carlo analysis. In contrast to Joint Probability Decision Making (JPDM), the proposed PJPDM technique does not require preemptive knowledge of weighting factors for competing objectives or goal constraints which can introduce bias into the final solution. Preemptive bias in a complex problem can degrade the overall capabilities of the final design. The implementation of PJPDM in this thesis eliminates the need for the numerical optimizer which is required with JPDM in order to improve upon a solution. In addition, a physics-based formulation is presented for the quantification of TMS safety effectiveness corresponding to debris impact/damage and how it can be applied towards risk mitigation. Lastly, a formulation loosely based on non-preemptive Goal Programming with equal weighted deviations is provided for the resolution of the inverse design space. This key step helps link vehicle capabilities to TMS technology subsystems in a top-down design approach. The methodology provides the designer more knowledge up front to help make proper engineering decisions and assumptions in the conceptual design phase regarding which technologies show greatest promise, and how to guide future technology research.

Diss. (sammanfattning) Umeå : Umeå universitet, 2009. / Härtill 5 uppsatser. Även tryckt utgåva.

Made available in DSpace on 2014-06-11T19:30:50Z (GMT). No. of bitstreams: 0 Previous issue date: 2010-02-18Bitstream added on 2014-06-13T20:07:53Z : No. of bitstreams: 1 matarucco_rr_dr_ilha.pdf: 2669428 bytes, checksum: c42a359efd2169b25baa0b6c3a967339 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Este trabalho apresenta dois métodos para a análise estática de contingências em Sistemas Elétricos de Potência utilizando o Método da Continuação. No primeiro método a margem de carregamento pós-contingência é obtida a partir do ponto de máximo carregamento do caso base. A magnitude de tensão de uma barra qualquer é usada como parâmetro na etapa de parametrização do fluxo de carga continuado. O ramo selecionado para avaliação da contingência é parametrizado por um fator de escalonamento que possibilita a remoção gradual do ramo e assegura a convergência nos casos em que o método diverge para a retirada total da linha de transmissão. Em geral, para a maioria das contingências analisadas são necessárias poucas iterações para a determinação do ponto de máximo carregamento pós-contingência. Mostra-se que o método pode ser usado como uma técnica alternativa para a averiguação e até mesmo para a obtenção da lista de contingências críticas fornecida pela função de análise de segurança de sistemas elétricos. No outro método, o qual obtém o ponto de máximo carregamento de pós-contingência a partir do caso base, as variáveis ângulo de fase e magnitude de tensão de uma barra k qualquer, e a perda total de potência ativa, são propostas como parâmetros para a etapa de parametrização do fluxo de carga continuado utilizado na averiguação da lista de contingências críticas fornecida pela função de análise de segurança de sistemas elétricos. Nos casos em que há divergência do fluxo de carga, o método proposto possibilita confirmar se esta ocorre devido à deficiência numérica do método em si ou a inexistência de um ponto de operação factível de pós-contingência. O uso da perda total de potência ativa como parâmetro traz como vantagem a possibilidade da determinação de pontos além do ponto de singularidade sem a necessidade... / This work presents two methods for static contingency analysis of electric power systems by using Continuation Methods. In the first the post-contingency loading margin is obtained from the base case maximum loading point. The voltage magnitude of any bus can be used as a parameter in the parameterization step of the proposed continuation power flow. The branch selected for contingency evaluation is parameterized by a scaling factor which allows its the gradual removal and assures the continuation power flow convergence for the cases where the method would diverge for the complete transmission line removal. In general, for most of the analyzed contingencies little iterations are necessary for the determination of the post-contingency maximum loading point. It is shown that the method can be used as an alternative technique to verify and even to obtain the list of critical contingencies supplied by the electric power systems security analysis function. In the other method, which obtains the maximum loading point from the base case, new parameters, namely the voltage magnitudes, phase angles and the total power losses, for evaluating the effects of branch outages. The approach can be used as a verification tool after a list of critical contingencies had been ranked according to their severities by the contingency selection functions. It is then possible to find whether the non-convergence of a power flow is due to a numerical problem or to an infeasible operating situation. The mains advantage of using the total real power losses as a parameter is that it is not necessary to change parameters during the solutions tracing until beyond the simple limit point, where the original Jacobian is singular. The proposed methods facilitate the development and the implementation of continuation methods for contingencies analysis

The central concern of this thesis is to describe the views of the young men in Feridun Zaimoglus book Kanak Sprak (2000) on their own identity ("Kanaken"), on the identity of other foreigners such as (college "Ali") and the Germans ("Alemannen") with help of close-reading. The book contains short-stories or "protocolls", written in the special language "KanakSprak", a mixture of Turkish and German street gang slang. There are claims that Zaimoglus texts are written based on hiphop/rap language. The categories in this work are identity, youth(sub)culture, territory, style and Hiphop. Other than close-reading the analysis will also take help of different research areas such as youth research and soliology.

La classification a été largement étudiée en apprentissage automatique. Les méthodes d’ensemble, qui construisent un modèle de classification en intégrant des composants d’apprentissage multiples, atteignent des performances plus élevées que celles d’un classifieur individuel. La précision de classification d’un ensemble est directement influencée par la qualité des données d’apprentissage utilisées. Cependant, les données du monde réel sont souvent affectées par les problèmes de bruit d’étiquetage et de déséquilibre des données. La marge d'ensemble est un concept clé en apprentissage d'ensemble. Elle a été utilisée aussi bien pour l'analyse théorique que pour la conception d'algorithmes d'apprentissage automatique. De nombreuses études ont montré que la performance de généralisation d'un classifieur ensembliste est liée à la distribution des marges de ses exemples d'apprentissage. Ce travail se focalise sur l'exploitation du concept de marge pour améliorer la qualité de l'échantillon d'apprentissage et ainsi augmenter la précision de classification de classifieurs sensibles au bruit, et pour concevoir des ensembles de classifieurs efficaces capables de gérer des données déséquilibrées. Une nouvelle définition de la marge d'ensemble est proposée. C'est une version non supervisée d'une marge d'ensemble populaire. En effet, elle ne requière pas d'étiquettes de classe. Les données d'apprentissage mal étiquetées sont un défi majeur pour la construction d'un classifieur robuste que ce soit un ensemble ou pas. Pour gérer le problème d'étiquetage, une méthode d'identification et d'élimination du bruit d'étiquetage utilisant la marge d'ensemble est proposée. Elle est basée sur un algorithme existant d'ordonnancement d'instances erronées selon un critère de marge. Cette méthode peut atteindre un taux élevé de détection des données mal étiquetées tout en maintenant un taux de fausses détections aussi bas que possible. Elle s'appuie sur les valeurs de marge des données mal classifiées, considérant quatre différentes marges d'ensemble, incluant la nouvelle marge proposée. Elle est étendue à la gestion de la correction du bruit d'étiquetage qui est un problème plus complexe. Les instances de faible marge sont plus importantes que les instances de forte marge pour la construction d'un classifieur fiable. Un nouvel algorithme, basé sur une fonction d'évaluation de l'importance des données, qui s'appuie encore sur la marge d'ensemble, est proposé pour traiter le problème de déséquilibre des données. Cette méthode est évaluée, en utilisant encore une fois quatre différentes marges d'ensemble, vis à vis de sa capacité à traiter le problème de déséquilibre des données, en particulier dans un contexte multi-classes. En télédétection, les erreurs d'étiquetage sont inévitables car les données d'apprentissage sont typiquement issues de mesures de terrain. Le déséquilibre des données d'apprentissage est un autre problème fréquent en télédétection. Les deux méthodes d'ensemble proposées, intégrant la définition de marge la plus pertinente face à chacun de ces deux problèmes majeurs affectant les données d'apprentissage, sont appliquées à la cartographie d'occupation du sol. / Classification has been widely studied in machine learning. Ensemble methods, which build a classification model by integrating multiple component learners, achieve higher performances than a single classifier. The classification accuracy of an ensemble is directly influenced by the quality of the training data used. However, real-world data often suffers from class noise and class imbalance problems. Ensemble margin is a key concept in ensemble learning. It has been applied to both the theoretical analysis and the design of machine learning algorithms. Several studies have shown that the generalization performance of an ensemble classifier is related to the distribution of its margins on the training examples. This work focuses on exploiting the margin concept to improve the quality of the training set and therefore to increase the classification accuracy of noise sensitive classifiers, and to design effective ensemble classifiers that can handle imbalanced datasets. A novel ensemble margin definition is proposed. It is an unsupervised version of a popular ensemble margin. Indeed, it does not involve the class labels. Mislabeled training data is a challenge to face in order to build a robust classifier whether it is an ensemble or not. To handle the mislabeling problem, we propose an ensemble margin-based class noise identification and elimination method based on an existing margin-based class noise ordering. This method can achieve a high mislabeled instance detection rate while keeping the false detection rate as low as possible. It relies on the margin values of misclassified data, considering four different ensemble margins, including the novel proposed margin. This method is extended to tackle the class noise correction which is a more challenging issue. The instances with low margins are more important than safe samples, which have high margins, for building a reliable classifier. A novel bagging algorithm based on a data importance evaluation function relying again on the ensemble margin is proposed to deal with the class imbalance problem. In our algorithm, the emphasis is placed on the lowest margin samples. This method is evaluated using again four different ensemble margins in addressing the imbalance problem especially on multi-class imbalanced data. In remote sensing, where training data are typically ground-based, mislabeled training data is inevitable. Imbalanced training data is another problem frequently encountered in remote sensing. Both proposed ensemble methods involving the best margin definition for handling these two major training data issues are applied to the mapping of land covers.

Resumo: Na região do Cerrado, a integração lavoura-pecuária sob sistema plantio direto é uma das alternativas para elevar a produtividade e o desempenho econômico de sistemas agrícolas. O trabalho de pesquisa objetivou: 1) avaliar a produtividade de grãos da cultura de milho em épocas de consorciação com o Panicum maximum cv. Tanzânia, Panicum maximum cv. Mombaça, Brachiaria brizantha cv. Marandu e Brachiaria ruziziensis; 2) avaliar a adubação nitrogenada dos capins após a colheita do milho em quatro épocas de corte no inverno/primavera quanto à produtividade de massa seca, a composição bromatológica, o índice ICF, a eficiência de conversão do N-fertilizante em forragem e a decomposição da palha após a última época de corte; 3) avaliar o efeito residual da adubação nitrogenada nos capins na cultura do milho em sucessão; 4) avaliar o desempenho econômico do milho consorciado, da pastagem adubada com nitrogênio, do milho em sucessão e do sistema como um todo, considerando o ganho de peso vivo médio diário de 0,2 e 0,6 kg/animal. Para atingir tais propósitos, foram conduzidos três experimentos sequenciais, durante os anos agrícolas de 2007/08 e 2008/09 na Fazenda de Ensino, Pesquisa e Extensão, da Faculdade de Engenharia - Unesp, Campus de Ilha Solteira, em um Latossolo Vermelho distroférrico em condições de cerrado, com histórico de cinco anos sob sistema plantio direto. O delineamento experimental utilizado foi o de blocos casualizados, com quatro repetições. No experimento I (2007/2008), os tratamentos foram constituídos de oito consórcios da cultura do milho com capins (semeado simultaneamente ou por ocasião da adubação nitrogenada de cobertura) e do milho sem consorciação. No expetimento II, após a colheita da cultura do milho (2007/2008), em esquema de parcelas subdivididas constituídas pela ausência e doses de 200, 400 e 800 kg/ha/ano de N na ... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: In the Brazilian "Cerrado" conditions, the crop-livestock integration under no-tillage system is an alternative to increase the yield and the economic performance of agricultural systems. The present work aimed: 1) to evaluate corn grain yield under different intercropping times with Panicum maximum cv. Tanzânia, Panicum maximum cv. Mombaça, Brachiaria brizantha cv. Marandu e Brachiaria ruziziensis; 2) to evaluate the nitrogen fertilization of the grasses after the corn harvest in four seasons cut in the winter/spring, considering the dry mass yield, chemical composition, ICF index, conversion efficiency of N-fertilizer on forage and the straw decomposition in "Litter-bags" after the last season of cut; 3) to evaluate the residual effect of nitrogen fertilization on grasses in corn crop in succession; 4) to evaluate to economic performance of corn intercropped, pasture fertilized with nitrogen, the corn crop in succession and the crop-livestock integration system as a whole, considering the daily weight gain of 0.2 and 0.6 kg/animal. To reach these purposes, they were conducted three sequenced studies during the 2007/2008 and 2008/2009 growing seasons, at Experimental Station from College of Engineering in Ilha Solteira, São Paulo State, Brazil, in a Red Latosol (Oxisol) in "Cerrado" conditions, with a history of five years under no-tillage system. The experimental design was a completely randomized block, with four replicates. In the experiment I (2007/2008), the treatments consisted of eigth methods for growing corn intercropped with Panicum maximum cv. Tanzania, Panicum maximum cv. Mombaça, Brachiaria brizantha cv. Marandu, and Brachiaria ruziziensis, sown simultaneously or at time of side dressing nitrogen fertilization, besides single grown corn. In the experiment II, after the corn harvest (2007/2008), in split-plot scheme was applied urea at side dressing ... (Complete abstract click electronic access below) / Orientador: Marcelo Andreotti / Coorientador: Antonio Fernando Bergamaschine / Banca: Enes Furlani Júnior / Banca: Ciniro Costa / Mestre