Made available in DSpace on 2015-12-03T18:35:23Z (GMT). No. of bitstreams: 1 Erasmo_Aparecido_Piccolo.pdf: 1376633 bytes, checksum: e1caf1219ae76bbc8c1e8dd1273f1aa4 (MD5) Previous issue date: 2012-09-05 / The court appointed expertise in the judicial exerts an important social function, to assist the court with explanations of technical and / or scientific, helping make decisions in the course of civil proceedings. This requires that the work of the expert, the expert report materialized accounting, is quality. The number of lawsuits related to financial institutions leads the top of the list of the Courts of Justice, which provided the opportunity to investigate the performance of professional experts in this environment. The eim of this study was to investigate the procedures used by experts appointed by the judges in the region of Bauru-SP in civil operations with overdraft. The research is descriptive, qualitative and documentary analysis. The raised the necessary procedures to work in the literature and expert in the rules of the CFC - Federal Accounting Council. In the sequel, began the field research, in an exploratory way, through the analysis documents obtained from six cases to civil courts in the region of Bauru-SP, in which we investigated the performanced of experts and sentences handed down by judges. The assessment was made by triangulation of existing procedures in the literature and in the CFC rules with the reports of the experts, accountants, administrators and economists, with the sentences handed down by judges. The model developed allowed gouge the quality of expert advice through compliance procedures and expert Degree Award of Use - GUL - by judges, with the use of keywords. The results showed that the accounting expertise is required in civil operations with overdraft and also the studies analyzed are good quality. It was found that increasing the quality and/or compliance with existing procedures in the literature and in the CFC rules are reflected in the GUL, causing its increase. The research revealed that the higher the quality of forensic work, the greater its use in their sentencing judges / A per??cia cont??bil no ??mbito judicial exerce uma fun????o social relevante, ao auxiliar o juiz com esclarecimentos de natureza t??cnica e / ou cient??fica, ajudando-o tomar decis??es no curso do processo civil. Isso requer que o trabalho do perito, materializado no laudo pericial cont??bil, seja de qualidade. A quantidade de a????es judiciais relacionadas ??s institui????es financeiras lidera o topo da lista dos Tribunais de Justi??a, o que ofereceu a oportunidade para investigar a atua????o profissional dos peritos neste ambiente. O objetivo do estudo foi investigar os procedimentos adotados por peritos nomeados pelos ju??zes na regi??o de Bauru-SP em processos c??veis de opera????es com cheque especial. A pesquisa ?? descritiva, qualitativa e com an??lise documental. Foram levantados os procedimentos necess??rios ao trabalho pericial existentes na literatura e nas normas do CFC - Conselho Federal de Contabilidade. Na sequencia, iniciou-se a pesquisa de campo, de forma explorat??ria, por meio da an??lise documental de seis processos obtidos junto ??s varas c??veis da regi??o de Bauru-SP, nos quais foram investigadas as atua????es dos peritos e as senten??as proferidas pelos ju??zes. A aprecia????o se deu pela triangula????o dos procedimentos existentes na literatura e nas normas do CFC com os laudos dos peritos, contadores, administradores e economistas, com as senten??as proferidas pelos ju??zes. O modelo elaborado permitiu aferir a qualidade do laudo pericial por meio do cumprimento dos procedimentos periciais e o Grau de Utiliza????o do Laudo - GUL - pelos ju??zes, com o uso de palavras-chave. Os resultados evidenciaram que a per??cia cont??bil ?? necess??ria nos processos c??veis de opera????es com cheque especial e, ainda, os trabalhos analisados s??o de boa qualidade. Constatou-se que o aumento da qualidade e/ou a concord??ncia dos procedimentos existentes na literatura e nas normas do CFC refletem-se no GUL, provocando seu aumento. A pesquisa revelou que, quanto maior a qualidade do trabalho pericial, maior a sua utiliza????o nas senten??as proferidas pelos ju??zes

Artificial neural networks have been widely used for machine learning and optimization. A neuro ensemble is a collection of neural networks that works cooperatively on a problem. In the literature, it has been shown that by combining several neural networks, the generalization of the overall system could be enhanced over the separate generalization ability of the individuals. Evolutionary computation can be used to search for a suitable architecture and weights for neural networks. When evolutionary computation is used to evolve a neuro ensemble, it is usually known as evolutionary neuro ensemble. In most real-world problems, we either know little about these problems or the problems are too complex to have a clear vision on how to decompose them by hand. Thus, it is usually desirable to have a method to automatically decompose a complex problem into a set of overlapping or non-overlapping sub-problems and assign one or more specialists (i.e. experts, learning machines) to each of these sub-problems. An important feature of neuro ensemble is automatic problem decomposition. Some neuro ensemble methods are able to generate networks, where each individual network is specialized on a unique sub-task such as mapping a subspace of the feature space. In real world problems, this is usually an important feature for a number of reasons including: (1) it provides an understanding of the decomposition nature of a problem; (2) if a problem changes, one can replace the network associated with the sub-space where the change occurs without affecting the overall ensemble; (3) if one network fails, the rest of the ensemble can still function in their sub-spaces; (4) if one learn the structure of one problem, it can potentially be transferred to other similar problems. In this thesis, I focus on classification problems and present a systematic study of a novel evolutionary neuro ensemble approach which I call cooperative coevolutionary mixture of experts (CCME). Cooperative coevolution (CC) is a branch of evolutionary computation where individuals in different populations cooperate to solve a problem and their fitness function is calculated based on their reciprocal interaction. The mixture of expert model (ME) is a neuro ensemble approach which can generate networks that are specialized on different sub-spaces in the feature space. By combining CC and ME, I have a powerful framework whereby it is able to automatically form the experts and train each of them. I show that the CCME method produces competitive results in terms of generalization ability without increasing the computational cost when compared to traditional training approaches. I also propose two different mechanisms for visualizing the resultant decomposition in high-dimensional feature spaces. The first mechanism is a simple one where data are grouped based on the specialization of each expert and a color-map of the data records is visualized. The second mechanism relies on principal component analysis to project the feature space onto lower dimensions, whereby decision boundaries generated by each expert are visualized through convex approximations. I also investigate the regularization effect of learning by forgetting on the proposed CCME. I show that learning by forgetting helps CCME to generate neuro ensembles of low structural complexity while maintaining their generalization abilities. Overall, the thesis presents an evolutionary neuro ensemble method whereby (1) the generated ensemble generalizes well; (2) it is able to automatically decompose the classification problem; and (3) it generates networks with small architectures.

Online learning algorithms have several key advantages compared to their batch learning algorithm counterparts: they are generally more memory efficient, and computationally mor efficient; they are simpler to implement; and they are able to adapt to changes where the learning model is time varying. Online algorithms because of their simplicity are very appealing to practitioners. his thesis investigates several online learning algorithms and their application. The thesis has an underlying theme of the idea of combining several simple algorithms to give better performance. In this thesis we investigate: combining weights, combining hypothesis, and (sort of) hierarchical combining.¶ Firstly, we propose a new online variant of the Bayes point machine (BPM), called the online Bayes point machine (OBPM). We study the theoretical and empirical performance of the OBPm algorithm. We show that the empirical performance of the OBPM algorithm is comparable with other large margin classifier methods such as the approximately large margin algorithm (ALMA) and methods which maximise the margin explicitly, like the support vector machine (SVM). The OBPM algorithm when used with a parallel architecture offers potential computational savings compared to ALMA. We compare the test error performance of the OBPM algorithm with other online algorithms: the Perceptron, the voted-Perceptron, and Bagging. We demonstrate that the combinationof the voted-Perceptron algorithm and the OBPM algorithm, called voted-OBPM algorithm has better test error performance than the voted-Perceptron and Bagging algorithms. We investigate the use of various online voting methods against the problem of ranking, and the problem of collaborative filtering of instances. We look at the application of online Bagging and OBPM algorithms to the telecommunications problem of channel equalization. We show that both online methods were successful at reducing the effect on the test error of label flipping and additive noise.¶ Secondly, we introduce a new mixture of experts algorithm, the fixed-share hierarchy (FSH) algorithm. The FSH algorithm is able to track the mixture of experts when the switching rate between the best experts may not be constant. We study the theoretical aspects of the FSH and the practical application of it to adaptive equalization. Using simulations we show that the FSH algorithm is able to track the best expert, or mixture of experts, in both the case where the switching rate is constant and the case where the switching rate is time varying.

De nombreuses catastrophes de diverses envergures frappent régulièrement des populations partout dans le monde. Parmi les exemples marquant on peut citer l'ouragan Katrina en 2005, le tremblement de terre en Haïti en 2010 ou plus récemment le Tsunami au Japon et la catastrophe de Fukujima qui a suivie. Au cours de ces catastrophes, plusieurs centaines d'organisations, comme la police, les pompiers ou les organisations d'aide humanitaire, interviennent pour sauver les gens et aider à revenir à une vie normale. Ces organisations ont besoin de se coordonner pour faire face à une situation dynamique avec des ressources limitées et une vision partielle de la situation. L'évolution de la situation entraîne souvent des changements d'objectif et de plan. Un des problèmes typique est d'obtenir un aperçu sur les relations entre ce qui a été fait, ce qui se passe actuellement et quelles sont les prochaines étapes. Ce problème est particulièrement difficile sur le plan inter-organisationnel : Chaque organisation coordonne la réponse de sa propre perspective et s'appuie sur les informations fournies par d'autres organisations. Notre objectif dans cette thèse est d'étudier comment supporter la coordination des activités par des personnes de différentes organisations dans une situation dynamique par un système d'information. L'idée de base est de tirer profit d'une approche basée sur les processus, où les activités et leurs relations sont rendues explicites. Nous présentons un cadre pour la coordination des activités dans des situations dynamiques. Il permet la modélisation ad hoc des relations entre ce qui a été fait, ce qui se passe actuellement et quelles sont les prochaines étapes. Les écarts par rapport au modèle et comment les activités ont été réalisées sont affichées à l'utilisateur pour mettre en évidence l'impact de l'évolution des objectifs. Nous étendons ce cadre au niveau inter-organisationnel. Certaines activités peuvent être partagées entre différentes organisations. Tout n'est pas partagé entre tout le monde pour tenir compte du respect de la vie privée, de la réglementation, des raisons stratégiques ou autres. Les activités partagées sont reproduites dans les espaces de travail de ces organisations. Nous décrivons comment des vues divergentes sur les activités et leurs relations peuvent etre détectées et traitées afin de revenir éventuellement à une vue convergente. Les concepts sont mis en œuvre comme une extension d'un service de collaboration distribuée ouvert. Ils ont été évalués par des gestionnaires de catastrophes expérimentés. Par ailleurs, nous avons conçu une expérience visant à évaluer l'utilisation d'outils pour aborder ces question. Nous avons effectué plusieurs expériences pour valider cette expérience. D'autres expériences pourront fournir une validation plus complété du modèle proposé dans cette thèse.

Drawing on the study of knowledge and action as a reference, this thesis explores how practitioners in two different firms perform their practice, 'knowing' and 'acting' simultaneously. It argues that types of knowledge, activities, individuals and context are interwoven at the moment of acting. However, this relationship is not static. The empirical work in a local labour placement company and in a multinational consultancy firm shows that practitioners resort to a host of different expressions of knowledge (or bundle of knowledge) when acting. Therefore, the prevailing role of explicit knowledge and the need for being a precedent in order to be applied is called into question.The empirical work also reveals that the bundle of knowledge is not static. It evolves over time and at the same time the prevailing type of knowledge varies depending on the type of practice and the practitioner's level of expertise. Therefore, the results underscore the fact that the relationship between knowledge and action is more dynamic and that both interplay simultaneously. Finally, this research shows that formal company categorisations (senior/junior) describe different practices rather than correspond to differential stocks of formal knowledge. This means that even when experts and novices apparently seem to be doing the same job, their actions are different as they are constituted through different combinations of knowledge types and orders of relevance. These results seem to point toward the fact that the essence of expertise resides in the expert's ability to reframe. He is able to reconstruct practice, whether by reframing his tasks or the overarching context.

L'école se préoccupe de la transmission de savoirs qui ne se réfèrent pas tous à des corps de savoirs savants constitués, comme ceux des mathématiques, par exemple. Il en est ainsi de la recherche d'information via Internet qui prend une place grandissante dans de nombreuses pratiques sociales. Dans cette perspective, un enseignement de la consultation d'information a été introduit dans les programmes de technologie de la classe de 4e. La thèse a pour objet d'examiner la question des savoirs à transmettre et de l'organisation des modes de transmission-appropriation de ce nouvel objet d'enseignement. La consultation d'information relève d'une activité humaine qui s'organise autour de l'usage, et donc de l'instrumentalisation d'un outil. Nous connaissons la faiblesse des apprentissages fondés sur la seule maîtrise opératoire. Examiner les enjeux de la transposition didactique à partir de pratiques, suppose donc de s'extraire d'une approche exclusivement procédurale, pour, dans un premier temps, examiner la question de l'élaboration de savoirs experts en nous appuyant sur la genèse instrumentale et à partir d'analyse de pratiques. Dans un second temps, nous nous sommes intéressés au processus d'enseignement apprentissage. L'observation d'élèves nous a permis, au travers de leurs conceptions, de mesurer la distance qui sépare leurs pratiques de consultation d'information de celles d'experts et ainsi de qualifier les enjeux de l'apprentissage. L'analyse du travail des enseignants permet d'identifier les modes de transmission qu'ils utilisent pour cet enseignement et donc d'anticiper sur les probables difficultés des élèves. Ainsi, le croisement de ces analyses nous permet de qualifier le processus de transmission-appropriation d'un corpus de savoirs particuliers liés à la consultation d'information via Internet et, au final, de penser des aides au travail des enseignants dans une perspective d'ingénierie didactique.

L'objectiu general d'aquest treball és trobar i mostrar una eina que permeti obtenir unarepresentació dels senyals procedents de sistemes dinàmics adequada a les necessitatsdels sistemes de Supervisió Experta de processos. Aquest objectiu general es pot subdividir en diverses parts, que són tractades en els diferents capítols que composen el treball i que es poden resumir en els següents punts:En primer lloc, cal conèixer les necessitats dels sistemes de Supervisió: La gran quantitat de dades que provenen dels processos fa necessari el tractament d'aquestes dades per obtenir-ne d'altres, més elaborades, amb un nivell més elevat de representació.La utilització de raonament qualitatiu, pròpia dels éssers humans, comporta la necessitat de representar simbòlicament els senyals, de traduir les dades numèriques en símbols.La Supervisió de sistemes dinàmics comporta que el temps sigui una variable fonamental, la asincronia dels esdeveniments significatius per a la Supervisió fa que les representacions més adequades i útils dels senyals siguinasíncrones. Finalment,l'ús dels coneixements experimentals en la Supervisió dels processos comporta que les representacions més naturals siguin les més útils.Aquestes necessitats fan de la representació dels senyals mitjançant episodis l'eina amb més possibilitats per assolir els objectius que es volen assolir. Per això, es presenta un formalisme que permet descriure i incloure-hi la formalització i les diferents aproximacions a aquest tipus de representació ja existents i, al mateix temps, augmentar-ne la significació a través de característiques dels senyals que noes tenen en compte en les aproximacions ja existents.El següent pas és aprofitar el nou formalisme per obtenir una nova representació amb un grau més gran de significació, cosa que s'aconsegueix representant explícitament les discontinuïtats i els períodes estacionaris o d'estabilitat, moltsignificatius en Supervisió de processos.Un problema sempre present en el tractament de senyals és el soroll que els afecta. Per aquest motiu es presenta un mètode que permet filtrar el soroll de manera que les representacions resultants quedin afectades el mínim possible per aquest tractament.Finalment, es presenta l'aplicació en línia de les eines descrites. La representació en línia dels senyals comporta el tractament de la incertesa inherent al coneixement parcial del senyal (un episodi no pot ser determinat i caracteritzat completament fins que no s'acaba). L'obtenció de resultats amb determinats graus de certesa és perfectament coherent amb la seva utilització posterior mitjançant Sistemes Expertso altres eines de la IA. Totes les aportacions del treball vénen acompanyades d'exemples i/o aplicacions que permeten observar-ne la utilitat i les limitacions.

This thesis is about ‘military manuals’ produced in the first few centuries of the Roman imperial period. It argues that these texts merit far more attention and appreciation than they have received in the scholarship so far. I will explore areas such as the way in which their authors order and rank Greek and Roman knowledge, engage with ideas about knowledge and power, help construct identity and discuss ethics and behavior. In the first chapter I will determine whether the authors operate within a specific ‘genre’, or ‘genres’, of military writing. Then I will explore how the texts relate to other traditions of technical texts, questions of audience, and finally the issue of their practicality. The second chapter will examine how authors tackle the issue of ‘Greek’ and ‘Roman’ knowledge, categorize, rank and use it for self-promotion. We will see how Roman knowledge is both subverted but also praised, and how Greek knowledge is at the same time placed above Roman knowledge and integrated into a narrative of continuity with it. The third chapter will focus on the use of Greek knowledge in the construction of Roman identity. I will explore how ‘manuals’ play a part in the identity of the Roman Empire, fitting into a picture of unity in diversity, and show how they contribute to Hadrian’s self-presentation. The fourth chapter will examine the ethical component in manuals. I will determine whether there was an ethical code of conduct in battle in the Classical world and whether it was different from general ethical norms. Then, we will examine whether our texts engage in any way with this ‘code’ and whether their individual approaches have anything in common or are fundamentally different.

La reconnaissance faciale (RF) est un domaine de recherche très actif en raison de ses nombreuses applications dans le domaine de la vision par ordinateur en général et en biométrie en particulier. Cet intérêt est motivé par plusieurs raisons. D’abord, le visage est universel. Ensuite, il est le moyen le plus naturel par les êtres humains de s’identifier les uns des autres. Enfin, le visage en tant que modalité biométrique est présente un caractère non intrusif, ce qui le distingue d’autres modalités biométriques comme l’iris ou l’emprunte digitale. La RF représente aussi des défis scientifiques importants. D’abord parce que tous les visages humains ont des configurations similaires. Ensuite, avec les images faciales 2D que l’on peut acquérir facilement, la variation intra-classe, due à des facteurs comme le changement de poses et de conditions d’éclairage, les variations d’expressions faciales, le vieillissement, est bien plus importante que la variation inter-classe.Avec l’arrivée des systèmes d’acquisition 3D capables de capturer la profondeur d’objets, la reconnaissance faciale 3D (RF 3D) a émergé comme une voie prometteuse pour traiter les deux problèmes non résolus en 2D, à savoir les variations de pose et d’éclairage. En effet, les caméras 3D délivrent généralement les scans 3D de visages avec leurs images de texture alignées. Une solution en RF 3D peut donc tirer parti d’une fusion avisée d’informations de forme en 3D et celles de texture en 2D. En effet, étant donné que les scans 3D de visage offrent à la fois les surfaces faciales pour la modalité 3D pure et les images de texture 2D alignées, le nombre de possibilités de fusion pour optimiser le taux de reconnaissance est donc considérable. L’optimisation de stratégies de fusion pour une meilleure RF 3D est l’objectif principal de nos travaux de recherche menés dans cette thèse.Dans l’état d’art, diverses stratégies de fusion ont été proposées pour la reconnaissance de visages 3D, allant de la fusion précoce "early fusion" opérant au niveau de caractéristiques à la fusion tardive "late fusion" sur les sorties de classifieurs, en passant par de nombreuses stratégies intermédiaires. Pour les stratégies de fusion tardive, nous distinguons encore des combinaisons en parallèle, en cascade ou multi-niveaux. Une exploration exhaustive d’un tel espace étant impossible, il faut donc recourir à des solutions heuristiques qui constituent nos démarches de base dans le cadre des travaux de cette thèse.En plus, en s’inscrivant dans un cadre de systèmes biométriques, les critères d’optimalité des stratégies de fusion restent des questions primordiales. En effet, une stratégie de fusion est dite optimisée si elle est capable d’intégrer et de tirer parti des différentes modalités et, plus largement, des différentes informations extraites lors du processus de reconnaissance quelque soit leur niveau d’abstraction et, par conséquent, de difficulté.Pour surmonter toutes ces difficultés et proposer une solution optimisée, notre démarche s’appuie d’une part sur l’apprentissage qui permet de qualifier sur des données d’entrainement les experts 2D ou 3D, selon des critères de performance comme ERR, et d’autre part l’utilisation de stratégie d’optimisation heuristique comme le recuit simulé qui permet d’optimiser les mélanges des experts à fusionner. [...] / Face recognition (FR) was one of the motivations of computer vision for a long time, but only in recent years reliable automatic face recognition has become a realistic target of biometrics research. This interest is motivated by several reasons. First, the face is one of the most preferable biometrics for person identification and verification related applications, because it is natural, non-intrusive, and socially well accepted. The second reason relates to the challenges encountered in the FR domain, in which all human faces are similar to each other and hence offer low distinctiveness as compared with other biometrics, e.g., fingerprints and irises. Furthermore, when employing facial texture images, intra-class variations due to various factors as illumination and pose changes are usually greater than inter-class ones, preventing 2D face recognition systems from being completely reliable in real conditions.Recent, 3D acquisition systems are capable to capture the shape information of objects. Thus, 3D face recognition (3D FR) has been extensively investigated by the research community to deal with the unsolved issues in 2D face recognition, i.e., illumination and pose changes. Indeed, 3D cameras generally deliver the 3D scans of faces with their aligned texture images. 3D FR can benefit from the fusion of 2D texture and 3D shape information.This Ph.D thesis is dedicated to the optimization of fusion strategies based on three dimensional data. However, there are some problems. Indeed, since the 3D face scans provide both the facial surfaces for the 3D model and 2D texture images, the number of fusion method is high.In the literature, many fusion strategies exist that have been proposed for 3D face recognition. We can roughly classify the fusion strategies into two categories: early fusion and late fusion. Some intermediate strategies such as serial fusion and multi-level fusion have been proposed as well. Meanwhile, the search for an optimal fusion scheme remains extraordinarily complex because the cardinality of the space of possible fusion strategies. It is exponentially proportional to the number of competing features and classifiers. Thus, we require fusion technique to efficiently manage all these features and classifiers that constitute our contribution in this work. In addition, the optimality criteria of fusion strategies remain critical issues. By definition, an optimal fusion strategy is able to integrate and take advantage from different data.To overcome all these difficulties and propose an optimized solution, we adopted the following reflection. [...]

The thesis deals with chances and difficulties, which a family bringing up a child with polio has to cope with. In the theoretical part a child with a polio is characterized, the influence of the birth and upbringing of a child with polio on the life of its siblings and parents is described. Next, the cooperation with experts, organizations and institutions is considered as well as help from the society in the form of special advantages and social benefits. In the practical part the experience, opinions and feelings of parents bringing up a child with polio are presented on the base of a performed research. The main objective of the research was to find out how the birth and upbringing of a child with polio influences the family life, which difficulties and chances in various forms the parents met with and what is their opinion of them.