公務人員的人事陞遷是一個複雜性極高，其中隱藏著許多不變的定律及過程，長官與部屬、各公務人員人之間的關係，更是如同蜘蛛網狀般的錯綜複雜，而各公務人員的陞遷狀況，更是隱藏著許多派系之間的鬥爭拉扯連動，或是提攜後進的過程，目前透過政府公開的總統府公報－總統令，可以清楚得知所有公務人員的任職相關資料，其中包含各職務之間的陞遷、任命、派免等相關資訊，而每筆資料亦包含機關、單位、職稱及職等資料，可以提供各種研究使用。 本篇係整理出一種陞遷序列的資料模型來進行研究，透過資料探勘的相關演算法－支撐向量機(Support Vector Machine，簡稱SVM)及決策樹(Decision Tree)的方式，並透過人事的領域知識加以找出較具影響力的屬性，來設計實驗的模型，並使用多組模型及多重資料進行實驗，透過整體平均預測結果及圖表方式來呈現各類別的預測狀況，再以不同的屬性資料來運算產生其相對結果，來分析其合理性，最後再依相關數據來評估此一方法的合理及可行性。 透過資料探勘設計的分類預測模型，其支撐向量機與決策樹都具有訓練量越大，展現之預測結果也愈佳之現象，這跟一般模型是相同的，而挖掘的主管職務屬性參數及關鍵屬性構想都跟人事陞遷的邏輯不謀而合，而預測結果雖各有所長，但整體來看則為支撐向量機略勝一籌，惟支撐向量機有一狀況，必須先行排除較不具影響力之屬性參數資料，否則其產生超平面的邏輯運算過程將產生拉扯作用，導致影響其預測結果；而決策樹則無是類狀況，且其應用較為廣泛，可以透過宣告各屬性值的類型，來進行不同屬性資料類型的分類實驗。 而透過支撐向量機與決策樹的產生的預測結果，其正確率為百分之77至82左右，如此顯示出國內中高階文官的陞遷制度是有脈絡可循的，其具有一定的制度規範及穩定性，而非隨意的任免陞遷；如此透過以上資料探勘的應用，藉著此特徵研究提供公務部門在進行人力資源管理、組織發展、陞遷發展以及組織部門精簡規劃上，作為調整設計參考的一些相關資訊；另透過一些相關屬性的輸入，可提供尚在服務的公務人員協助其預估陞遷發展的狀況，以提供其進行相關生涯規劃。 / The employee promotion is a highly complexity task in Government office, it include many invariable laws and the process, between the senior officer and the subordinate, various relationships with other government employees, It’s the similar complex with the spider lattice, and it hides many clique's struggles in Government official’s promotion, and help to process the promote for the junior generation, through the government public presidential palace - presidential order, it‘s able to get clearly information about all government employees’ correlation data, include various related information like promotion, recruitment , and each data also contains the instruction, like the job unit, job title and job rank for all research reference. It organizes a promoted material model to conduct the research, by the material exploration's related calculating method – Support Vector Machine (SVM) and the decision tree, and through by knowledge of human resource to discover the influence to design the experiment's model, and uses the multi-group models and materials to process, and by this way , it can get various categories result by overall average forecasting and the graph, then operates by different attribute material to get relative result and analyzes its rationality, finally it depends on the correlation data to re-evaluate its method reasonable and feasibility. To this classification forecast model design, the SVM and the decision tree got better performance together with the good training quality, it’s the same with the general model, and it’s the same view to find the details job description for senior management and employee promotion, however the forecasting result has their own strong points, but for the totally, the SVM is slightly better, only if any accidents occurred, it needs to elimination the attribute parameter material which is not have the big influence, otherwise it will have the planoid logic operation process to produce resist status, and will affect its forecasting result, but the decision tree does not have this problem, and its application is more widespread, it can through by different type to make the different experiment. The forecasting result through by SVM and decision tree, its correction percentage can be achieved around 77% - 82% , so it indicated the high position level promotion policy should be have its own rules to follow, it has certain system standard and the stability, but non-optional promoted, so trough by the above data mining, follow by this characteristic to provide Government office to do the Human resource management, organization development, employee promotion and simplify planning to the organization, takes the re-design information for reference, In addition through by some related attribute input, it may provide the government employee who is still on duty and assist them to evaluate promotion development for future career plan.

Cis-regulatory elements are the short regions of DNA to which specific regulatory proteins bind and these interactions subsequently influence the level of transcription for associated genes, by inhibiting or enhancing the transcription process. It is known that much of the genetic change underlying morphological evolution takes place in these regions, rather than in the coding regions of genes. Identifying these sites in a genome is a non-trivial problem. Experimental (wet-lab) methods for finding binding sites exist, but all have some limitations regarding their applicability, accuracy, availability or cost. On the other hand computational methods for predicting the position of binding sites are less expensive and faster. Unfortunately, however, these algorithms perform rather poorly, some missing most binding sites and others over-predicting their presence. The aim of this thesis is to develop and improve computational approaches for the prediction of transcription factor binding sites (TFBSs) by integrating the results of computational algorithms and other sources of complementary biological evidence. Previous related work involved the use of machine learning algorithms for integrating predictions of TFBSs, with particular emphasis on the use of the Support Vector Machine (SVM). This thesis has built upon, extended and considerably improved this earlier work. Data from two organisms was used here. Firstly the relatively simple genome of yeast was used. In yeast, the binding sites are fairly well characterised and they are normally located near the genes that they regulate. The techniques used on the yeast genome were also tested on the more complex genome of the mouse. It is known that the regulatory mechanisms of the eukaryotic species, mouse, is considerably more complex and it was therefore interesting to investigate the techniques described here on such an organism. The initial results were however not particularly encouraging: although a small improvement on the base algorithms could be obtained, the predictions were still of low quality. This was the case for both the yeast and mouse genomes. However, when the negatively labeled vectors in the training set were changed, a substantial improvement in performance was observed. The first change was to choose regions in the mouse genome a long way (distal) from a gene over 4000 base pairs away - as regions not containing binding sites. This produced a major improvement in performance. The second change was simply to use randomised training vectors, which contained no meaningful biological information, as the negative class. This gave some improvement over the yeast genome, but had a very substantial benefit for the mouse data, considerably improving on the aforementioned distal negative training data. In fact the resulting classifier was finding over 80% of the binding sites in the test set and moreover 80% of the predictions were correct. The final experiment used an updated version of the yeast dataset, using more state of the art algorithms and more recent TFBSs annotation data. Here it was found that using randomised or distal negative examples once again gave very good results, comparable to the results obtained on the mouse genome. Another source of negative data was tried for this yeast data, namely using vectors taken from intronic regions. Interestingly this gave the best results.

Social anxiety disorder (SAD) is a common psychiatric disorder associated with considerable suffering. Cognitive behaviour therapy (CBT) has been shown to be effective but a significant proportion does not respond or relapses, stressing the need of augmenting treatment. Using neuroimaging could elucidate the psychological and neurobiological interaction and may help to improve current therapeutics. To address this issue, functional and structural magnetic resonance imaging (MRI) were repeatedly conducted on individuals with SAD randomised to receive CBT or an active control condition. MRI was performed pre-, and post-treatment, as well as at one-year follow-up. Matched healthy controls were also scanned to be able to evaluate disorder-specific neural responsivity and structural morphology. This thesis aimed at answering three major questions. I) Does the brain’s fear circuitry (e.g., the amygdala) change, with regard to neural response and structural morphology, immediately after CBT? II) Are the immediate changes in the brain still present at long-term follow-up? III) Can neural responsivity in the fear circuitry predict long-term treatment outcome at the level of the individual? Thus, different analytic methods were performed. Firstly, multimodal neuroimaging addressed questions on concomitant changes in neural response and grey matter volume. Secondly, two different experimental functional MRI tasks captured both neural response to emotional faces and self-referential criticism. Thirdly, support vector machine learning (SVM) was used to evaluate neural predictors at the level of the individual. Amygdala responsivity to self-referential criticism was found to be elevated in individuals with SAD, as compared to matched healthy controls, and the neural response was attenuated after effective CBT. In individuals with SAD, amygdala grey matter volume was positively correlated with symptoms of anticipatory speech anxiety, and CBT-induced symptom reduction was associated with decreased grey matter volume of the amygdala. Also, CBT-induced reduction of amygdala grey matter volume was evident both at short- and long-term follow-up. In contrast, the amygdala neural response was weakened immediately after treatment, but not at one-year follow-up. In extension to treatment effects on the brain, pre-treatment connectivity between the amygdala and the dorsal anterior cingulate cortex (dACC) was stronger in long-term CBT non-responders, as compared to long-term CBT responders. Importantly, by use of an SVM algorithm, pre-treatment neural response to self-referential criticism in the dACC accurately predicted (&gt;90%) the clinical response to CBT. In conclusion, modifying the amygdala is a likely mechanism of action in CBT, underlying the anxiolytic effects of this treatment, and the brain’s neural activity during self-referential criticism may be an accurate and clinically relevant predictor of the long-term response to CBT. Along these lines, neuroimaging is a vital tool in clinical psychiatry that could potentially improve clinical decision-making based on an individual’s neural characteristics. / Social ångest är en av de vanligaste psykiska sjukdomarna. Mer än en miljon svenskar bedöms lida av detta. Social ångest leder ofta till svåra konsekvenser för den som drabbas, men även ökade kostnader för samhället har noterats, t ex i form av ökad sjukfrånvaro. Även om många som drabbas inte söker hjälp så finns effektiva behandlingar för social ångest, både farmakologiska och psykologiska behandlingar rekommenderas av Socialstyrelsen. Kognitiv beteendeterapi (KBT) är en evidensbaserad och rekommenderad psykologisk behandling för social ångest. Trots att nuvarande interventioner är effektiva så är det fortfarande en andel individer som inte blir förbättrade. Det finns en stor andel studier som visar att individer med social ångest, i jämförelse med friska individer, karakteriseras av överdriven aktivitet i ett nätverk som har till uppgift att tolka och reagera på hotfull information. Denna aktivitet är lokaliserad i rädslonätverket där området amygdala spelar en central roll. Det finns ett behov att utveckla nuvarande behandlingar och denna avhandling syftar till att öka vår förståelse för en neurobiologisk verkningsmekanism bakom KBT för social ångest. I detta forskningsprojekt har magnetresonanstomografi (MRT) använts för att undersöka personer som lider av social ångest. Upprepade mätningar har genomförts, innan, efter, och vid uppföljning ett år efter ångestlindrande behandling. Utöver detta har individer som inte lider av social ångest undersökts för att förstå hur patienter skiljer sig från friska personer, men också för att undersöka om behandlingen normaliserar patientens hjärna. Under tiden som deltagarna undersöktes med MRT genomfördes två experiment för att ta reda på hur hjärnan reagerar på affektiv information. Deltagarna tittade på bilder med ansikten som uttrycker emotioner, t ex arga och rädda ansiktsuttryck, samt information som innehöll kritiska kommentarer riktade till personen själv eller någon annan, t ex ”ingen tycker om dig” eller ”hon är inkompetent”. Strukturella bilder på deltagarnas hjärnor har också samlats in vid varje mättillfälle. Utöver detta fick alla deltagare instruktioner om att de efter MRT skulle hålla en muntlig presentation inför en publik. Denna uppgift är oftast den värsta tänkbara för individer med social ångest, och syftet med uppgiften var att relatera hjärnans struktur och aktivitet till hur mycket ångest som individerna upplevde inför denna situation. I arbetet med denna avhandling har tre frågor ställts. a) Uppstår strukturella och funktionella förändringar i rädslonätverket direkt efter avslutad KBT (Studie I och II)? b) Är de tidiga förändringarna efter behandlingen även kvarstående ett år senare (Studie III)? c) Kan hjärnans reaktioner i rädslonätverket förutspå vilka individer som kommer att bli förbättrade av en ångestlindrande psykologisk behandling på lång sikt? Resultat från studierna i denna avhandling sammanfattas nedan: Reaktioner till självriktad kritik i amygdala är överdrivna hos individer med social ångest, i jämförelse med friska individer Reaktioner i amygdala minskar efter att individerna blivit behandlade med KBT och minskningarna korrelerar till minskade symptom av social ångest Den strukturella volymen av amygdala korrelerar positivt med hur mycket ångest individerna upplever inför en muntlig presentation, och minskningen av dessa symptom korrelerar även med hur mycket volymen av amygdala minskar efter KBT Minskningen av amygdalavolym och den samtidigt minskade reaktiviteten i amygdala till självriktad kritik är korrelerade. Medieringsanalyser antyder att det är den minskade volymen som driver förhållandet mellan minskad reaktivitet och minskad ångest inför att hålla en muntlig presentation Den strukturella minskningen av amygdala ses både direkt efter behandlingens avslut, men även vid uppföljning ett år senare. Hjärnans reaktivitet till självriktad kritik i amygdala minskar direkt efter behandling, men är inte kvarstående vid uppföljning ett år senare Kopplingen mellan hjärnans reaktivitet till självriktad kritik i amygdala och dorsala främre cingulum var starkare hos de som inte blev förbättrade (jämfört med de som blev bättre) av en ångestlindrande behandling på lång sikt Med hjälp av en stödvektormaskin (en. support vector machine learning) och ett mönster av hjärnaktivitet i dorsala främre cingulum innan behandling påbörjades, predicerades (med 92% träffsäkerhet) vilka individer som ett år senare var fortsatt förbättrade av en effektiv psykologisk behandling Utifrån dessa observationer är slutsatserna att strukturell och funktionell påverkan på amygdala är en möjlig neurobiologisk mekanism för minskad social ångest efter KBT, samt att reaktivitet i främre cingulum kan ge kliniskt relevant data om vem som kommer att bli förbättrad av en psykologisk behandling. Denna information kan potentiellt vara viktig i framtidens psykiatri för att utveckla existerande behandlingar, men även för att stödja klinikers beslutsfattande huruvida en viss individ bör erbjudas en specifik behandling eller ej. / <p>Illustration on the cover by Jan Lööf. Cover image printed with permission from Jan Lööf and Bonnier Carlsen Förlag. The cover was art directed by Staffan Lager.</p><p>The thesis is reprinted and the previous ISBN was 9789176856888.</p>

La Voix sur IP (VoIP) est devenue un paradigme majeur pour fournir des services de télécommunications flexibles tout en réduisant les coûts opérationnels. Le déploiement à large échelle de la VoIP est soutenu par l'accès haut débit à l'Internet et par la standardisation des protocoles dédiés. Cependant, la VoIP doit également faire face à plusieurs risques comprenant des vulnérabilités héritées de la couche IP auxquelles s'ajoutent des vulnérabilités spécifiques. Notre objectif est de concevoir, implanter et valider de nouveaux modèles et architectures pour assurer une défense préventive, permettre le monitorage et la détection d'intrusion dans les réseaux VoIP. Notre travail combine deux domaines: celui de la sécurité des réseaux et celui de l'intelligence artificielle. Nous renforçons les mécanismes de sécurité existants en apportant des contributions sur trois axes : Une approche basée sur des mécanismes d'apprentissage pour le monitorage de trafic de signalisation VoIP, un pot de miel spécifique, et un modèle de corrélation des événements pour la détection d'intrusion. Pour l'évaluation de nos solutions, nous avons développés des agents VoIP distribués et gérés par une entité centrale. Nous avons développé un outil d'analyse des traces réseaux de la signalisation que nous avons utilisé pour expérimenter avec des traces de monde réel. Enfin, nous avons implanté un prototype de détection d'intrusion basé sur des règles de corrélation des événements. / Voice over IP (VoIP) has become a major paradigm for providing flexible telecommunication services and reducing operational costs. The large-scale deployment of VoIP has been leveraged by the high-speed broadband access to the Internet and the standardization of dedicated protocols. However, VoIP faces multiple security issues including vulnerabilities inherited from the IP layer as well as specific ones. Our objective is to design, implement and validate new models and architectures for performing proactive defense, monitoring and intrusion detection in VoIP networks. Our work combines two domains: network security and artificial intelligence. We reinforce existent security mechanisms by working on three axes: a machine learning approach for VoIP signaling traffic monitoring, a VoIP specific honeypot and a security event correlation model for intrusion detection. In order to experiment our solutions, we have developed VoIP agents which are distributed and managed by a central entity. We have developed an analyzer of signaling network traces and we used it to analyze real-world traces. Finally, we have implemented a prototype of a rule-based event-driven intrusion detection system.

The project revolves around logistics and its adoption to the new rules. Theobjective of this project is to focus on minimizing data tampering to the lowest level possible.To achieve the set goals in this project, Decision support system and simulation havebeen used. However, to get clear insight about how they can be implemented, a systematicliterature review (Case Study incl.) has been conducted, followed by interviews with personnelat Kakinada port to understand the real-time complications in the field. Then, a simulatedexperiment using real-time data from Kakinada port has been conducted to achieve the set goalsand improve the level of transparency on all sides i.e., shipper, port and terminal.

Dans cette thèse, nous abordons le problème de la segmentation des images de documents en proposant de nouvelles approches pour la détection et la classification de leurs contenus. Dans un premier lieu, nous étudions le problème de l'estimation d'inclinaison des documents numérisées. Le but de ce travail étant de développer une approche automatique en mesure d'estimer l'angle d'inclinaison du texte dans les images de document. Notre méthode est basée sur la méthode Maximum Gradient Difference (MGD), la R-signature et la transformée de Ridgelets. Nous proposons ensuite une approche hybride pour la segmentation des documents. Nous décrivons notre descripteur de trait qui permet de détecter les composantes de texte en se basant sur la squeletisation. La méthode est appliquée pour la segmentation des images de documents numérisés (journaux et magazines) qui contiennent du texte, des lignes et des régions de photos. Le dernier volet de la thèse est consacré à la détection du texte dans les photos et posters. Pour cela, nous proposons un ensemble de descripteurs de texte basés sur les caractéristiques du trait. Notre approche commence par l'extraction et la sélection des candidats de caractères de texte. Deux méthodes ont été établies pour regrouper les caractères d'une même ligne de texte (mot ou phrase) ; l'une consiste à parcourir en profondeur un graphe, l'autre consiste à établir un critère de stabilité d'une région de texte. Enfin, les résultats sont affinés en classant les candidats de texte en régions « texte » et « non-texte » en utilisant une version à noyau du classifieur Support Vector Machine (K-SVM) / In this thesis I discuss the document image segmentation problem and I describe our new approaches for detecting and classifying document contents. First, I discuss our skew angle estimation approach. The aim of this approach is to develop an automatic approach able to estimate, with precision, the skew angle of text in document images. Our method is based on Maximum Gradient Difference (MGD) and R-signature. Then, I describe our second method based on Ridgelet transform.Our second contribution consists in a new hybrid page segmentation approach. I first describe our stroke-based descriptor that allows detecting text and line candidates using the skeleton of the binarized document image. Then, an active contour model is applied to segment the rest of the image into photo and background regions. Finally, text candidates are clustered using mean-shift analysis technique according to their corresponding sizes. The method is applied for segmenting scanned document images (newspapers and magazines) that contain text, lines and photo regions. Finally, I describe our stroke-based text extraction method. Our approach begins by extracting connected components and selecting text character candidates over the CIE LCH color space using the Histogram of Oriented Gradients (HOG) correlation coefficients in order to detect low contrasted regions. The text region candidates are clustered using two different approaches ; a depth first search approach over a graph, and a stable text line criterion. Finally, the resulted regions are refined by classifying the text line candidates into « text» and « non-text » regions using a Kernel Support Vector Machine K-SVM classifier

L’avantage du train depuis sa création est sa faible résistance à l’avancement du fait du contact fer-fer de la roue sur le rail conduisant à une adhérence réduite. Cependant cette adhérence faible est aussi un inconvénient majeur : étant dépendante des conditions environnementales, elle est facilement altérée lors d’une pollution du rail (végétaux, corps gras, eau, etc.). Aujourd’hui, les mesures prises face à des situations d'adhérence dégradée impactent directement les performances du système et conduisent notamment à une perte de capacité de transport. L’objectif du projet est d’utiliser les nouvelles technologies d’imagerie spectrale pour identifier sur les rails les zones à adhérence réduite et leur cause afin d’alerter et d’adapter rapidement les comportements. La stratégie d’étude a pris en compte les trois points suivants : • Le système de détection, installé à bord de trains commerciaux, doit être indépendant du train. • La détection et l’identification ne doivent pas interagir avec la pollution pour ne pas rendre la mesure obsolète. Pour ce faire le principe d’un Contrôle Non Destructif est retenu. • La technologie d’imagerie spectrale permet de travailler à la fois dans le domaine spatial (mesure de distance, détection d’objet) et dans le domaine fréquentiel (détection et reconnaissance de matériaux par analyse de signatures spectrales). Dans le temps imparti des trois ans de thèse, nous nous sommes focalisés sur la validation du concept par des études et analyses en laboratoire, réalisables dans les locaux de SNCF Ingénierie & Projets. Les étapes clés ont été la réalisation d’un banc d’évaluation et le choix du système de vision, la création d'une bibliothèque de signatures spectrales de référence et le développement d'algorithmes classification supervisées et non supervisées des pixels. Ces travaux ont été valorisés par le dépôt d'un brevet et la publication d'articles dans des conférences IEEE. / The advantage of the train since its creation is in its low resistance to the motion, due to the contact iron-iron of the wheel on the rail leading to low adherence. However this low adherence is also a major drawback : being dependent on the environmental conditions, it is easily deteriorated when the rail is polluted (vegetation, grease, water, etc). Nowadays, strategies to face a deteriorated adherence impact the performance of the system and lead to a loss of transport capacity. The objective of the project is to use a new spectral imaging technology to identify on the rails areas with reduced adherence and their cause in order to quickly alert and adapt the train's behaviour. The study’s strategy took into account the three following points : -The detection system, installed on board of commercial trains, must be independent of the train. - The detection and identification process should not interact with pollution in order to keep the measurements unbiased. To do so, we chose a Non Destructive Control method. - Spectral imaging technology makes it possible to work with both spatial information (distance’s measurement, target detection) and spectral information (material detection and recognition by analysis of spectral signatures). In the assigned time, we focused on the validation of the concept by studies and analyses in laboratory, workable in the office at SNCF Ingénierie & Projets. The key steps were the creation of the concept's evaluation bench and the choice of a Vision system, the creation of a library containing reference spectral signatures and the development of supervised and unsupervised pixels classification. A patent describing the method and process has been filed and published.

Made available in DSpace on 2017-07-21T14:19:22Z (GMT). No. of bitstreams: 1 Sandro Teixeira.pdf: 611887 bytes, checksum: da75c60dae366a84db89509883f57db4 (MD5) Previous issue date: 2014-07-31 / Considered a quality indicator, carbon constitutes an important attribute in the productive capacity of the soil. However the traditional methodologies used for determining carbon cause environmental problems due to the use of chemical reagents. The replacement of this procedure by others that generate little or no amount of toxic waste has been considered important. Spectroscopy is one of the promising techniques in Precision Agriculture for soil analysis and can be used to estimate carbon content. Among its benefits, highlights the sample preservation, no consumption of reagents, and their efficiency acquiring data from a large number of samples. The aim of this work was to contribute to determine a regression model able to predict the carbon content in soil samples using spectroscopy in the visible and near infrared region. The Machine Learning SVM technique available in the WEKA software was used to create the model. Because of their generalization ability SVM has been considered a better alternative than the other methods of multivariate regression. Two sets of soil samples collected in the Campos Gerais region were used to the experiments. The results evaluation was based on the forecast errors and the correlation coefficients between the values carbon content predicted by the model. Correlation coefficients ranging from 0.84 to 0.90 were found. It was concluded that the NIRS-vis spectroscopy combined with SVM technique can be recommended as an alternative to conventional methods for carbon analysis in the soil. / Considerado um indicador de qualidade, o carbono constitui-se em um importante atributo na capacidade produtiva do solo. Porém, as tradicionais metodologias empregadas para sua determinação geram problemas ambientais devido ao uso de reagentes químicos. Diante disso, a substituição desse procedimento por outros que gerem menor ou nenhuma quantidade de resíduos tóxicos tem sido considerada relevante. A espectroscopia é uma das técnicas promissora na Agricultura de Precisão para análises de solos e que pode trazer uma solução viável para análise de teor de carbono. Dentre suas vantagens, destaca-se a preservação da amostra, o não consumo de reagentes, além de sua eficiência na aquisição de dados provenientes de um grande número de amostras. O objetivo deste trabalho foi contribuir com um modelo de regressão capaz de predizer a quantidade de carbono em amostras de solo utilizando a espectroscopia na região do visível e no infravermelho próximo. Para tanto, foi utilizada a técnica de Aprendizagem de Máquina SVM incorporada ao software WEKA como auxílio na criação do modelo. A SVM tem representado uma alternativa melhor aos já consagrados métodos de regressão multivariada por apresentar capacidade de generalização. Nos experimentos realizados foram utilizados dois conjuntos de amostras de solo coletadas na região dos Campos Gerais. A avaliação dos resultados teve como base os erros de previsão e os coeficientes de correlação entre os valores dos teores de carbono preditos pelo modelo. Foram encontrados coeficientes de correlação que variaram entre 0,84 a 0,90. Concluiu-se que a espectroscopia no vis-NIRS aliada à técnica SVM é recomendada como uma alternativa aos métodos convencionais de análise de carbono em solos.

La reconnaissance des perturbations survenant sur les réseaux HTA est une problématique essentielle pour les clients industriels comme pour le gestionnaire du réseau. Ces travaux de thèse ont permis de développer un système d’identification automatique. Il s’appuie sur des méthodes de segmentation qui décomposent de manière précise et efficace les régimes transitoires et permanents des perturbations. Elles utilisent des filtres de types Kalman linéaire ou anti-harmoniques pour extraire les régimes transitoires. La prise en compte des variations harmoniques et de la présence de transitoires proches se fait à l’aide de seuils adaptatifs. Des méthodes de correction du retard a posteriori permettent d’améliorer la précision de la décomposition. Des indicateurs adaptés à la dynamique des régimes de fonctionnement analysés sont utilisés pour caractériser les perturbations. Peu sensibles aux erreurs de segmentation et aux perturbations harmoniques, ils permettent une description fiable des phases des perturbations. Deux types de systèmes de décision ont également été étudiés : des systèmes experts et des classifieurs SVM. Ces systèmes ont été mis au point à partir d’une large base de perturbations simulées. Leurs performances ont été évaluées sur une base de perturbations réelles : ils déterminent efficacement le type et la direction des perturbations observées (taux de reconnaissance moyen > 98%). / The recognition of disturbances affecting MV networks is essential to industrials and distribution system operators. The aim of this thesis work is to design a near real-time automatic system able to detect and identify disturbances from their waveforms. Segmentation methods split the disturbed waveforms into transient and steady-state intervals. They use Kalman filters or anti-harmonic filters to extract the transient intervals. Adaptive thresholding methods increase the detection capacity while a posterior delay compensation methods improve the accuracy of the decomposition. Indicators adapted to the disturbance dynamic are used to characterize its steady-state and transient phases. They are robust to segmentation inaccuracies as well as to steady-state disturbances such as harmonics. Two distinct decision systems are also studied: expert recognition systems and SVM classifiers. During the learning stage, a large simulated event database is used to train both systems. Their performances are evaluated on real events: the type and direction of the measured disturbances are determined with a recognition rate over 98%.

經濟體的產出缺口一直是政策執行者的觀察重點，當一國出現產出缺口時，代表資源配置並不均衡，將發生通貨膨脹或是失業的現象，如能提早預期到未來是否會出現產出缺口，將可讓政策執行者即早進行政策實施，且有文獻指出，殖利率曲線資料中具有隱含未來經濟狀況之資訊。 本研究以美國財政部與聯準會之公開資料，將以殖利率曲線之斜率進行預測產出缺口；本文研究美國1977年至2016年之國民生產毛額成分與殖利率之資料，目標為建立對於未來一季將出現正向或負向缺口現象之模型，本研究建立三種預測模型進行比較，分別為線性迴歸模型、羅吉斯迴歸模型與機器學習中的支持向量機，以實質GDP的缺口預測而言，研究結果顯示，三者預測準確度均達到65%以上，支持向量機的準確度更達到80.85%。 得出以下結論，第一，殖利率曲線對於未來總體經濟產出缺口具有一定之解釋力；第二，對於高維度之預測模型在機器學習中的支持向量機表現會較一般常用之迴歸模型佳；第三，進出口的預測力在三個模型下均表現較差，可能為殖利率曲線對於進出口並不具有完整有效的資訊，可能有其餘的經濟指標或金融市場資訊可以解釋；第四，對於實質消費與投資等民間部門經濟行為有超過80%的預測力。 / The output gap of the economy has always been the objectives of policy practitioners. When a country appear the output gap, it means that the allocation of resources is not equilibrium and the inflation or unemployment will occur. The output gap will allow policymakers to implement the policy as early as possible, and the literature notes that the information of the yield curve has information about the future economic situation. In this paper, we using the data from the U.S. Department of Treasury and the Federal Reserve to predict the output gap by the slopes of the yield curve. Our goal is to construct the prediction model for the next quarter. To forecast the real GDP gap, three prediction models were compared, linear regression model, logistic regression model and support vector machine. The results show that the accuracy of the three predictions are more than 65%, support vector machine accuracy to reach 80.85%. We can have conclusions showing below: First, the yield curve has significant explanatory power for the overall economic output gap in the future. Second, the support vector machine perform better than the commonly used regression model. Third, the predictive power of real import and export in the three models are poor performance, there may be the rest of the economic indicators or financial market information can be explained. Fourth, the real consumption and investment has the predictive power more than 80% of the forecast.