Dans ce travail, nous exploitons des propriétés déjà connues pour les systèmes de poids des représentations afin de les définir pour les orbites des groupes de Weyl des algèbres de Lie simples, traitées individuellement, et nous étendons certaines de ces propriétés aux orbites des groupes de Coxeter non cristallographiques. D'abord, nous considérons les points d'une orbite d'un groupe de Coxeter fini G comme les sommets d'un polytope (G-polytope) centré à l'origine d'un espace euclidien réel à n dimensions. Nous introduisons les produits et les puissances symétrisées de G-polytopes et nous en décrivons la décomposition en des sommes de G-polytopes. Plusieurs invariants des G-polytopes sont présentés. Ensuite, les orbites des groupes de Weyl des algèbres de Lie simples de tous types sont réduites en l'union d'orbites des groupes de Weyl des sous-algèbres réductives maximales de l'algèbre. Nous listons les matrices qui transforment les points des orbites de l'algèbre en des points des orbites des sous-algèbres pour tous les cas n<=8 ainsi que pour plusieurs séries infinies des paires d'algèbre-sous-algèbre. De nombreux exemples de règles de branchement sont présentés. Finalement, nous fournissons une nouvelle description, uniforme et complète, des centralisateurs des sous-groupes réguliers maximaux des groupes de Lie simples de tous types et de tous rangs. Nous présentons des formules explicites pour l'action de tels centralisateurs sur les représentations irréductibles des algèbres de Lie simples et montrons qu'elles peuvent être utilisées dans le calcul des règles de branchement impliquant ces sous-algèbres. / In this work, we exploit properties well known for weight systems of representations to define them for individual orbits of the Weyl groups of simple Lie algebras, and we extend some of these properties to orbits of non-crystallographic Coxeter groups. Points of an orbit of a finite Coxeter group G are considered as vertices of a polytope (G-polytope) centered at the origin of a real n-dimensional Euclidean space. Products and symmetrized powers of G-polytopes are introduced and their decomposition into the sums of G-polytopes is described. Several invariants of G-polytopes are found. The orbits of Weyl groups of simple Lie algebras of all types are reduced to the union of orbits of the Weyl groups of maximal reductive subalgebras of the algebra. Matrices transforming points of the orbits of the algebra into points of subalgebra orbits are listed for all cases n<=8 and for many infinite series of algebra-subalgebra pairs. Numerous examples of branching rules are shown. Finally, we present a new, uniform and comprehensive description of centralizers of the maximal regular subgroups in compact simple Lie groups of all types and ranks. Explicit formulas for the action of such centralizers on irreducible representations of the simple Lie algebras are given and shown to have application to computation of the branching rules with respect to these subalgebras.

(in English) This thesis explores relationships among the tenses that express the past in the English language. Among these tenses are: past simple, past continuous, present perfect, present perfect continuous, past perfect, and past perfect continuous. The research focuses on the variation between the past simple (which also includes past continuous) and the present perfect (which also includes progressive constructions). The researched variation is the use of the past simple in the context of the present perfect in which the Present-Day English (PDE) would use the present perfect, and vice versa. Three decades (1731-1740, 1791-1800, and 1861-1870) were chosen from the eighteenth and the nineteenth centuries. The findings were compared to the PDE situation. The material was collected from an online database called The Proceeding of the Old Bailey, which is believed to be one of the most reliable sources that are representative of the spoken language of the day. It was thought that the variation would be observed best in spontaneous spoken language. The aim was to study spoken language that was influenced by the grammatical prescriptivism of the age only marginally. There was an expectation that the occurrence of the past simple in the context of the present perfect would gradually decrease...

Wie schätzen Menschen quantitative Größen wie zum Beispiel den Verkaufspreis eines Autos? Oft benutzen Menschen zur Lösung von Schätzproblemen sogenannte Cues, Informationen, die probabilistisch mit dem zu schätzenden Kriterium verknüpft sind. Um den Verkaufspreis eines Autos zu schätzen, könnte man zum Beispiel Informationen über das Baujahr, die Automarke, oder den Kilometerstand des Autos verwenden. Um menschliche Schätzprozesse zu beschreiben, werden häufig linear additive Modelle herangezogen. In meiner Dissertation schlage ich alternative ein heuristisches Modell zur Schätzung quantitativer Größen vor: das Mapping-Modell. Im ersten Kapitel meiner Dissertation teste ich das Mapping-Modell gegen weitere, in der Literatur etablierte, Schätzmodelle. Es zeigte sich, dass das Mapping-Modell unter unterschiedlichen Bedingungen in der Lage war, die Schätzungen der Untersuchungsteilnehmer akkurat vorherzusagen. Allerdings bestimmte die Struktur der Aufgabe - im Einklang mit dem Ansatz der „adaptiven Werkzeugkiste“ - im großen Maße, welches Modell am besten geeignet war, die Schätzungen zu erfassen. Im zweiten Kapitel meiner Dissertation greife ich diesen Ansatz auf und untersuche, in wie weit die Aufgabenstruktur bestimmt, welches Modell die Schätzprozesse am Besten beschreibt. Meine Ergebnisse zeigten, dass das Mapping-Modell am Besten dazu geeignet war die Schätzungen der Versuchsteilnehmer zu beschreiben, wenn explizites Wissen über die Aufgabe vorhanden war, während ein Exemplar-Modell den Schätzprozess erfasste, wenn die Abstraktion von Wissen schwierig war. Im dritten Kapitel meiner Dissertation, wende ich das Mapping-Modell auf juristische Entscheidungen an. Eine Analyse von Strafakten ergab, dass das Mapping-Modell Strafzumessungsvorschläge von Staatsanwälten besser vorhersagte als eine lineare Regression. Dies zeigt, dass das Mapping-Modell auch außerhalb von Forschungslaboratorien dazu geeignet ist menschliche Schätzprozesse zu beschreiben. / How do people make quantitative estimations, such as estimating a car’s selling price? Often people rely on cues, information that is probabilistically related to the quantity they are estimating. For instance, to estimate the selling price of a car they could use information, such as the car’s manufacturer, age, mileage, or general condition. Traditionally, linear regression type models have been employed to capture the estimation process. In my dissertation, I propose an alternative cognitive theory for quantitative estimation: The mapping model which offers a heuristic approach to quantitative estimations. In the first part of my dissertation l test the mapping model against established alternative models of estimation, namely, linear regression, an exemplar model, and a simple estimation heuristic. The mapping model provided a valid account of people’s estimates outperforming the other models in a variety of conditions. Consistent with the “adaptive toolbox” approach on decision, which model was best in predicting participants’ estimations was a function of the task environment. In the second part of my dissertation, I examined further how different task features affect the performance of the models make. My results indicate that explicit knowledge about the cues is decisive. When knowledge about the cues was available, the mapping model was the best model; however, if knowledge about the task was difficult to abstract, participants’ estimations were best described by the exemplar model. In the third part of my dissertation, I applied the mapping model in the field of legal decision making. In an analysis of fining and incarceration decisions, I showed that the prosecutions’ sentence recommendations were better captured by the mapping model than by legal policy modeled with a linear regression. These results indicated that the mapping model is a valid model which can be applied to model actual estimation processes outside of the laboratory.

In dieser Dissertationsschrift soll erkl{\"a}rt werden, wie auf der Grundlage von einfachen Strata, wie sie in einer gemeinsamen Arbeit mit Broussous \cite{BG} betrachtet wurden, einfache Charaktere f{\"u}r lokale einfache Algebren konstruiert werden k{\"o}nnen, wobei die Konstruktion den Vorbildern von Bushnell und Kutzko im zerfallenden Fall \cite{BK1} und von Zink \cite{Z7} im Falle eines Schiefk{\"o}rpers folgt. Der Begriff des einfachen Charakters geht auf die Arbeit \cite{BK1} zur{\"u}ck und bezeichnet eine ausgezeichnete Auswahl von Heisenbergcharakteren, die zu einem stabilen Darstellungsfilter geh{\"o}ren, der gem{\"a}{ss} \cite{Z2}(Hauptsatz 1.4) einem Darstellungsfilter zugeordnet wird, der zu einer absteigenden Normalreihe $$1+\R\supset1+\R^2\supset\ldots$$ geh{\"o}rt, wobei $\R$ das Jacobsonradikal einer erblichen Ordnung bezeichnet. Wir werden hier nur von Hauptordnungen ausgehen, d.h. von dem Fall, da{ss} $\R$ und seine Potenzen gebrochene Hauptideale sind. Diese Vorgehensweise und auch die besondere Auswahl der Heisenbergcharaktere in Form von einfachen Charakteren, wird durch die Konstruktion im Falle eines Schiefk{\"o}rpers \cite{Z7} und durch den abstrakten Matchingsatz \cite{BDKV} gerechtfertigt. Im Falle eines lokalen zentralen Schiefk{\"o}rpers ist n{\"a}mlich der Bewertungsring die einzige erbliche Ordnung und die einfachen Charaktere sind alle Heisenbergcharaktere die zu einem stabilen Darstellungsfilter geh{\"o}ren, der gem{\"a}{ss} \cite{Z2}(Hauptsatz 1.4) einem Darstellungsfilter, der zur absteigenden Normalreihe $$1+\pin_D\supset1+\pin_D^2\supset\ldots$$ geh{\"o}rt, zugeordnet wird, wobei $\pin_D$ das Bewertungsideal des Schiefk{\"o}rpers $D$ bezeichnet. Der abstrakte Matchingsatz liefert nun die Existenz einer Bijektion zwischen den irreduziblen glatten Darstellungen der multiplikativen Gruppe des lokalen zentralen Schiefk{\"o}rpers $D$ und den irreduziblen quadratintegrierbaren glatten Darstellungen einer beliebigen anderen lokalen zentraleinfachen Algebra vom selben reduzierten Grad {\"u}ber demselben nicht-archimedischen Grundk{\"o}rper $F$, welche den Charakter einer Darstellung in dem Sinne erh{\"a}lt, da{ss} die Charakterwerte auf den Konjugationsklassen elliptischer Elemente der verschiedenen Algebren, welche mithilfe ihrer Minimalpolynome identifiziert werden k{\"o}nnen, bis auf ein Vorzeichen {\"u}bereinstimmen. Wir werden hier kanonische Bijektionen zwischen den einfachen Charakteren f{\"u}r verschiedene zentraleinfache Algebren vom selben reduzierten Grad {\"u}ber demselben Grundk{\"o}rper angeben, von denen wir erwarten, da{ss} sie mit der Abbildung des abstrakten Matchingsatzes vertr{\"a}glich sind. Das dieses in der Tat der Fall ist, wurde bisher nur in einfachen F{\"a}llen wie \cite{He} und \cite{BH2} gezeigt, jedoch wurde in der Arbeit \cite{Z4} bereits mithilfe der Konstruktionen von \cite{Z7} und \cite{BK3} eine Bijektion zwischen den irreduziblen glatten Darstellungen der multiplikativen Gruppe des lokalen zentralen Schiefk{\"o}rpers $D$ vom Index $N$ {\"u}ber einem Grundk{\"o}rper $F$ und den irreduziblen essentiell quadratintegrierbaren glatten Darstellungen von $Gl_N(F)$ konstruiert, welche den Artinf{\"u}hrer und den formalen Grad einer Darstellung erh{\"a}lt. Da die Abbildung des abstrakten Matchingsatzes dieselben Forderungen erf{\"u}llt, kommt dies der gew{\"u}nschten Vertr{\"a}glichkeit schon sehr nahe und wir erf{\"u}llen mit unserer Konstruktion insbesondere die in der Arbeit \cite{Z4} gemachte Forderung die dort im Bezug auf die einfachen Charaktere getroffen Auswahlen noch unabh{\"a}ngiger von den jeweiligen Algebren zu gestalten. Die hier getroffene Auswahl wird durch die Verwendung sogenannter spezieller approximierender Folgen getroffen, welche sich aus einer Verallgemeinerung der in \cite{BG} gemachten {\"U}berlegungen ergeben. Im Anschlu{ss} an die Konstruktion und den Vergleich einfacher Charaktere werden wir in einer gro{ss}en Anzahl von F{\"a}llen zeigen, da{ss} sich die Heisenbergdarstellungen, die wir zu den einfachen Charakteren erhalten, in kanonischer Weise fortsetzen lassen und wir erwarten von diesen Fortsetzungen, da{ss} sie analoge Eigenschaften besitzen, wie die sogenannten ``$\beta$-Fortsetzungen'' von \cite{BK1}(5.2.1) im zerfallenden Fall. Damit k{\"o}nnen wir in diesen F{\"a}llen eine Liste von hypothetischen einfachen Typen angeben, von denen wir vermuten, da{ss} sie alle Bernsteinkomponenten parametrisieren, welche irreduzible essentiell quadratintegrierbare Darstellungen enthalten. Insbesondere vermuten wir, da{ss} sich die supercuspidalen Darstellungen mittels kompakter Induktion aus Fortsetzungen solcher einfacher Typen auf eine kompakt modulo Zentrum Untergruppe gewinnen lassen. Um die Vollst{\"a}ndigkeit dieser Konstruktion zu demonstrieren, h{\"a}tten wir allerdings noch die Eigenschaft ``Verkettung impliziert Konjugation'' zu zeigen, welche wir ebenfalls auf eine Folgearbeit verschieben m{\"u}ssen. Beabsichtig w{\"a}re dann ein Vollst{\"a}ndigkeitsbeweis mit dem abstrakten Matchingsatz wie bei L. Corwin \cite{Co} oder in \cite{Z4}. Wir weisen hier nur in Spezialf{\"a}llen nach, dass die Typendarstellungen, welche wir hier angegeben haben, tats{\"a}chlich Typen im Sinne von \cite{BK4}(4.1)(4.2) sind. Insbesondere sind es auch unsere Berechnungen in der Arbeit \cite{GSZ}, welche dem von uns im Geiste von \cite{Z7} und \cite{BK1} gemachten Ans{\"a}tzen hohe Evidenz geben. / In this thesis, we try to explain how simple characters for arbitrary central simple algebras over a non-archimedian local field $F$ can be constructed. Moreover, we introduce a kind of matching of simple characters between different algebras of fixed reduced degree. If the index of the algebra $A$ is odd or $A=M_l(D)$, where $l$ is an arbitrary prime number and $D$ a central division algebra over $F$, we can extend the Heisenberg representations associated to the simple characters to level-0 and obtain a hypothetical list of simple types. For $A=M_l(D)$ and if the residual field of $F$ is not the field with two elements, we can proof that all so-called maximal simple types in our list are simple types in the sense of \cite{BK4} and their extensions to their stabelizers induce supercupidal representations of $G_l(D)$. Using the the heuristical relation via the abstract matching theorem of \cite{BDKV} to the cases of a division algebra due to \cite{Z5} and to the split case due to \cite{BK1}, we conjecture that all supercuspidal representations of $Gl_l(D)$ can be obtained by this way.

Cette thèse propose de nouvelles méthodes d'analyse d'enregistrements cérébraux intra-crâniens (potentiels de champs locaux), qui pallie les lacunes de la méthode temps-fréquence standard d'analyse des perturbations spectrales événementielles : le calcul d'une moyenne sur les enregistrements et l'emploi de l'activité dans la période pré-stimulus. La première méthode proposée repose sur la détection de sous-ensembles d'électrodes dont l'activité présente des synchronisations cooccurrentes en un même point du plan temps-fréquence, à l'aide de modèles bayésiens de mélange gaussiens. Les sous-ensembles d'électrodes pertinents sont validés par une mesure de stabilité calculée entre les résultats obtenus sur les différents enregistrements. Pour la seconde méthode proposée, le constat qu'un bruit blanc dans le domaine temporel se transforme en bruit ricien dans le domaine de l'amplitude d'une transformée temps-fréquence a permis de mettre au point une segmentation du signal de chaque enregistrement dans chaque bande de fréquence en deux niveaux possibles, haut ou bas, à l'aide de modèles bayésiens de mélange ricien à deux composantes. À partir de ces deux niveaux, une analyse statistique permet de détecter des régions temps-fréquence plus ou moins actives. Pour développer le modèle bayésien de mélange ricien, de nouveaux algorithmes d'inférence bayésienne variationnelle ont été créés pour les distributions de Rice et de mélange ricien. Les performances des nouvelles méthodes ont été évaluées sur des données artificielles et sur des données expérimentales enregistrées sur des singes. Il ressort que les nouvelles méthodes génèrent moins de faux-positifs et sont plus robustes à l'absence de données dans la période pré-stimulus / This thesis promotes new methods to analyze intracranial cerebral signals (local field potentials), which overcome limitations of the standard time-frequency method of event-related spectral perturbations analysis: averaging over the trials and relying on the activity in the pre-stimulus period. The first proposed method is based on the detection of sub-networks of electrodes whose activity presents cooccurring synchronisations at a same point of the time-frequency plan, using bayesian gaussian mixture models. The relevant sub-networks are validated with a stability measure computed over the results obtained from different trials. For the second proposed method, the fact that a white noise in the temporal domain is transformed into a rician noise in the amplitude domain of a time-frequency transform made possible the development of a segmentation of the signal in each frequency band of each trial into two possible levels, a high one and a low one, using bayesian rician mixture models with two components. From these two levels, a statistical analysis can detect time-frequency regions more or less active. To develop the bayesian rician mixture model, new algorithms of variational bayesian inference have been created for the Rice distribution and the rician mixture distribution. Performances of the new methods have been evaluated on artificial data and experimental data recorded on monkeys. It appears that the new methods generate less false positive results and are more robust to a lack of data in the pre-stimulus period

Ce travail porte sur différentes méthodes de détection de protéines par des anticorps (Acs) marqués par des boîtes quantiques (QD, « quantum dots »). La première partie de la thèse porte sur le développement et l'optimisation de méthodes de conjugaison d'Acs polyclonaux à des QDs. Une étude de 2007 avait démontré que les conjugués anticorps-quantum dots (Acs-QDs) commerciaux ne présentent que de très rares anticorps fonctionnels à leur surface [Pathak et al., 2007]. Nous avons donc : (i) développé des conditions de réduction ménagée des ponts disulfures des Acs en utilisant soit du dithiothreitol, soit du 2-mercapotethanolamine, (ii) purifié les fragments d'Acs fonctionnels grâce à une colonne d'affinité, (iii) conjugué ces fragments fonctionnels d'Acs aux QDs en utilisant l'agent de liaison amine-thiol SMCC et (iv) développé une méthode de purification des conjugués sur gel d'agarose afin d'éliminer les fragments d'Acs non conjugués et les QDs libres. Des tests en cytométrie en flux ont permis de déterminer l'efficacité des conjugués pour détecter l'expression de la molécule CD4 à la surface de lymphocytes T humains. Un marquage de CD4 réalisé avec des conjugués préparés selon notre méthode s'avère être cinq fois plus sensible qu'en utilisant des conjugués réalisés selon les recommandations commerciales. Une autre méthode de préparation des Acs pour la conjugaison aux QDs a été testée : l'ajout de groupements fonctionnels sulfhydriles sur des amines primaires grâce au N-succinimidyl S-acetylthioacetate (SATA). Cependant, cette méthode ne permet pas d'avoir un contrôle précis du nombre ni de l'emplacement des groupements fonctionnels sulfhydriles ainsi ajoutés. Les tests de conjugaison aux QDs qui ont été réalisés avec ces Acs-SH, en utilisant l'agent de liaison SMCC, a entraîné la formation d'agrégats de taille variable. Cette méthode a donc été abandonnée.La seconde partie de la thèse porte sur la démonstration de la faisabilité et l'optimisation d'un marquage de l'antigène carcino-embryonnaire (CEA, « carcinoembryonic antigen ») humain à la surface de lignées cellulaires murines avec un anticorps simple domaine (sdAb) anti-CEA biotinylé détecté grâce à des conjugués streptavidine-QDs commerciaux et analysé par cytométrie en flux. Ces sdAbs ont été biotinylés selon deux approches : (i) avec des agents de biotinylation chimique qui permettent d'ajouter une biotine sur les amines primaires (biotinylation in vitro) et (ii) par une enzyme lors de leur production par E. coli (biotinylation in vivo). La détection du CEA humain à la surface des 100 000 cellules (lignées cellulaires murine MC38 et MC38-CEA) par ces sdAbs biotinylés a été testée en cytométrie de flux puis comparée à celle obtenue par un anticorps monoclonal biotinylé in vitro. Les résultats ont démontré que les sdAbs biotinylés ont une sensibilité de détection similaire que la biotinylation soit réalisée in vitro ou in vivo. Par contre, la sensibilité de la détection du CEA est environ cinquante fois meilleure en utilisant des sdAbs biotinylés (0,6 fmol d'anticorps sont nécessaires pour détecter le CEA) qu'avec des anticorps monoclonaux biotinylés (33 fmol). / We have been working on different ways to detect proteins with antibodies (Abs) labeled by quantum dots (QDs). The first part of his work is to develop and optimize the conjugation of polyclonal antibodies to quantum dot. A study has reported that commercial antibody-quantum dots conjugates (Abs-QDs) present very few functional Ab fragments at the surface of the conjugate [Pathak et al., 2007]. We had: (i) developed an advanced procedure of antibody reduction using dithiothreitol (DTT) or 2-mercaptoethanolamine (2-MEA) (to prevent the loss of antibody functions), (ii) purified the active fragments by affinity purification on column, (iii) conjugated the active reduced antibody fragments to QDs using the amine-thiol crosslinker SMCC for SH coupling, and (iv) developed the purification of the conjugates on agarose gel to remove free QDs and unconjugated antibody fragments. Our conjugates present about a five times better ability to detect CD4 by flow cytometry on 500 000 isolated human lymphocyte T cells than those made after the commercial procedure. Another procedure for antibody preparation was addition of sulfhydryl groups on primary amines using N-succinimidyl S-acetylthioacetate (SATA). However this procedure presents some variable yield and the number and the localization of sulfhydryl groups added on antibody cannot be fully controlled. Thereby, the conjugation of these Abs-SH to QDs using SMCC chemistry leads to the creation of aggregates. This Ab preparation in preparation for conjugation to QDs was abandoned.The second part of our work focusses on testing and comparing the ability to detect carcinoembryonic antigen (CEA) by flow cytometry with anti-CEA single domain antibodies (sdAbs) labeled by QDs. The sdAbs were biotinylated by using two methods: (i) in vitro chemical biotinylation reagent which adds biotins on primary amines, and (ii) enzymatic in vivo biotinylation during their production in E. coli. These anti-CEA sdAb biotinylation methods were compared to in vitro biotinylated monoclonal Abs against CEA for their respective ability to detect human CEA on 100 000 mice cells (MC38 and MC38-CEA cell lines) using flow cytometry. The results show that the limit detection for biotinylated sdAbs is similar between in vitro and in vivo biotinylation. Furthermore the limit detection with biotinylated sdAb (0.6 fmol are required to detect CEA) is about fifty times better than with biotinylated monoclonal Abs (33 fmol).

No description available.

Made available in DSpace on 2017-07-21T18:53:09Z (GMT). No. of bitstreams: 1 Mary Dias Queji.pdf: 909281 bytes, checksum: 0364cd27c13b09019850a0944168a898 (MD5) Previous issue date: 2008-04-05 / The agro industry of the apple generates, during the processing, the pomace which is considered the main by-product. Studies show that of the total amount of fruit that is processed to obtain the apple juice, 20 to 40 correspond to this by-product, which, is usually destined as a complement in animal feed or delivered onto the soil as organic fertilizer. The chemical composition, in moist base, is constituted by moisture (80), fibers (5) and soluble solids (14) represented, mainly, by fructose, glucose and sucrose, as well as, by organic acids, represented, in the most part by malic acid. The apple pomace also presents phenolic compounds that are target nutrients of great sensory and nutritional importance. Therefore, the quantification of its constituents represents an important source of data in the characterization of apple pomace for biotechnological purposes, seeking to attribute an appropriate use for this by-product. The conventional methodologies used for the quantification of sugars, organic acids and total phenols, although being part of the routine analyses in the quality control laboratories, are onerous, time consuming and generate residues. The aim of this study was to develop a fast, versatile analytical technique, of low cost and no pollutant, using the diffuse reflectance infrared spectroscopy (DRIFTS) allied to methods of multivariate calibration (PLSR). For the construction of the multivariate models, the averages of the concentrations of the simple sugars, malic acid and total phenols were used, obtained by the conventional methodologies, as well as the data in the medium infrared spectroscopy (MID) and near (NIR), obtained in duplicate, and of the 52 spectra obtained for the samples of apple pomace, 47 made part of the set calibration and 5 of the set validation. The regression models for the prediction of the concentration of fructose, sucrose, total phenols and malic acid obtained better results in MID, with averages of relative standard errors of 3.9 (with 5 Latents Variable), 6.6 (with 5 Latents Variable), 6.4 (with 4 Latents Variable) and 5.9 (with 5 Latents Variable), respectively. Already the best capacity prediction for glucose concentration was obtained by NIR, in which the average of relative standard error was 7.4 (with 6 Latents Variable). The obtained results demonstrate the good capacity of prediction of the multivariate models based in infrared spectroscopy and characteristic advantages of the association DRIFTPLSR. / A agroindústria da maçã gera, durante o processamento, o bagaço que é considerado o principal subproduto deste setor. Levantamentos mostram que da quantidade total de fruta que é processada para a obtenção do suco de maçã, 20 a 40 correspondem a este subproduto, que, normalmente, é destinado como complemento na alimentação animal ou dispensado no solo como adubo orgânico. Sua composição química, em base úmida, é constituída por umidade (80), fibras (5) e sólidos solúveis (14), estes últimos representados, principalmente, por frutose, glucose e sacarose, bem como, ácidos orgânicos, representados, majoritariamente pelo ácido málico. O bagaço da maçã também apresenta compostos fenólicos, que são fito-nutrientes de grande importância sensorial e nutricional. Portanto, a quantificação desses constituintes representa uma importante fonte de dados na caracterização do bagaço de maçã para finalidades biotecnológicas, visando atribuir um fim mais nobre a este subproduto. As metodologias convencionais empregadas para a quantificação de açúcares, ácidos orgânicos e fenóis totais, embora façam parte das análises de rotina nos laboratórios de controle de qualidade, são onerosas, demoradas e geram resíduos. O objetivo deste trabalho foi desenvolver uma técnica analítica rápida, versátil, de baixo custo e não poluente. Para tal, utilizou-se a espectroscopia no infravermelho por refletância difusa (DRIFTS) aliada a métodos de calibração multivariada (Regressão de Mínimos Quadrados Parciais - PLSR). Para a construção dos modelos multivariados, foram utilizadas as médias das concentrações dos açúcares simples, ácido málico e fenóis totais, obtidas pelas metodologias convencionais, bem como, os dados de espectroscopia no infravermelho médio (MID) e próximo (NIR). Os espectros foram obtidos em duplicata, sendo que dos 52 espectros das amostras de bagaço de maçã, 47 fizeram parte do conjunto de calibração e 5 do conjunto de validação externa. Os modelos de regressão para a previsão da concentração de frutose, sacarose, fenóis totais e ácido málico obtiveram melhores resultados no MID, com médias de erro padrão relativo de 3.9 (com 5 Variáveis Latentes), 6.6 (com 5 Variáveis Latentes), 6.4 (com 4 Variáveis Latentes) e 5.9 (com 5 Variáveis Latentes), respectivamente. Já a melhor capacidade de previsão para a concentração de glucose foi obtida pelo NIR, na qual a média de erro padrão relativo foi de 7.4 (com 6 Variáveis Latentes). Os resultados obtidos demonstram a boa capacidade de previsão dos modelos multivariados fundamentados em espectroscopia no infravermelho e vantagens características da associação DRIFT-PLSR.

Made available in DSpace on 2016-04-29T13:17:58Z (GMT). No. of bitstreams: 1 MariaPaulaMontans.pdf: 976390 bytes, checksum: 501a51586442d57ea969b712b2f444c6 (MD5) Previous issue date: 2006-05-10 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Stimulus and response classes are defined by a given set of shared properties. Stimulus classes are identified when different stimuli affect an organism in the same way. Processes, procedures and variables responsible for the establishment of stimulus control and stimulus classes and for the testing of such classes have been studied and today a number of authors argue that such processes and procedures may generate distinct types of stimulus classes such as functional stimulus classes and equivalent stimulus classes. Matching to sample is nowadays the choice procedure to establish conditional discriminations and classes of equivalent stimuli. Nevertheless, other procedures for the establishment of conditional discriminations among stimuli that prove to be effective for the emergence of equivalent classes have been described. The current study tested if a procedure that established errorless (or almost errorless) simple simultaneous discriminations based on discriminations already established, without any reversals, could establish conditional discriminations and equivalence classes formed by a SD series and a SΔ series of stimuli. Six children aged between 6 and 7 participated in the study. They worked in a computer equipped with a programme that controlled all the experiment conditions and also recorded the necessary data. Six experimental conditions, were successively presented: (1) simultaneous discrimination training of 3 stimuli pairs named A, B, C, (2) discriminative training of all stimuli pairs with 100% and 50% of trials reinforced, (3) conditional discrimination training, (4) test of emergent stimuli relations, (5) training and test of stimuli class formation, (6) grouping card stimuli. Results have shown that each participant has achieved criterion on the simple discrimination training of stimuli pairs A, B and C without errors or almost without errors or reversals and that the procedure allowed the establishment not only of the simple discriminations among the 3 pairs of stimuli, but also aloowed for the emergence of conditional relations among the S+ and S- stimuli series, for three participants / Classes de estímulos têm sido definidas por um conjunto de determinadas propriedades que compartilham. Analistas do comportamento têm estudado procedimentos para produzir e para identificar ou avaliar a formação de classes de estímulos e mais recentemente têm proposto que há mais de um tipo de classe de estímulos, por exemplo, classes de estímulos funcionais (ou funcionalmente equivalentes) e classes de estímulos equivalentes. Tais classes se distinguiriam pelas relações que os estímulos pertencentes a uma mesma classe estabelecem entre si e/ou pelos seus efeitos sobre o responder. Nestes casos, os procedimentos envolvidos no estabelecimento de classes de estímulos seriam responsáveis pela formação de distintas classes e de distintos tipos de classes. O procedimento de matching to sample, por exemplo, tem sido visto como procedimento privilegiado para estabelecer discriminações condicionais entre estímulos e para testar se houve o estabelecimento de relações entre estímulos tais que as três propriedades definidoras de classes de estímulos equivalentes emergem. No entanto, outros procedimentos de treino discriminativo também têm sido apontados como procedimentos efetivos para o estabelecimento de classes de estímulos tais que se pode afirmar a existência das relações entre os estímulos que são definidoras de classes de estímulos equivalentes. O presente trabalho teve por objetivo estudar se um procedimento que estabelece conjuntos de discriminações simultâneas simples, com base em discriminações já existentes, sem reversões repetidas e sem (ou quase sem) erros, possibilitaria a emergência de relações condicionais entre uma série de estímulos SD e entre uma série de SΔ. Seis crianças entre 6 e 7 anos de idade participaram deste estudo. Os participantes trabalharam em um computador equipado com um programa que controlava as condições do experimento e registrava as respostas. O presente experimento teve um pré-treino e seis fases: (1) treino discriminativo de 3 pares de estímulos denominados A, B e C, (2) treino discriminativo dos 3 pares mixados com 100% e 50% das tentativas reforçadas, (3) treino de discriminação condicional, (4) teste de relações emergentes entre estímulos, (5) treino e teste de formação de classes de estímulos, e (6) agrupamento de estímulos em cartões. Os resultados mostraram que todos os participantes aprenderam as discriminações simples entre os pares A, B e C sem erro ou quase sem erro. Foi constatado também que esse procedimento de discriminação simples, sem reversões ou erros permitiu estabelecer discriminações simples entre os 3 pares de estímulos e permitiu também a emergência de relações condicionais entre as séries de estímulos SD e entre a série de SΔ para três participantes. Palavras chaves: discriminação simples, discriminação condicional, classes de estímulos, classes de estímulos funcionais, classes de estímulos equivalentes

Made available in DSpace on 2016-04-29T13:18:02Z (GMT). No. of bitstreams: 1 Daniel Carvalho.pdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2007-04-04 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The present research had two goals: (a) to test the efficacy of a delayed prompt procedure to develop errorless discrimination between letters and their rotations, and (b) to test the effects of the discrimination training when MTS trials with the trained letters, theirs rotations, and similar letters were presented. Participated in the study 9 children (aged 3 to 6) who did not have a perfect performance when discrimination between pairs of graphically similar letters and their rotations (b, d, n, u, p, q, and their 270o rotations, and the letter B, D, N, U, P, Q and their 180o rotations). A test with identity-matching-to-sample trials involving the S+ letters as sample stimuli and the same letters, their rotations and similar letters as comparison stimuli was first conducted. A discrimination training of each testes letter (S+) and its rotation (S_) was then conducted: During training, colors (previously established as S+ and S-) were superimposed on the new S+ (letter) and S- (letter-rotation) with longer delays after each successful trial, until the child systematically responded choosing S+ before the color was presented. After training the initial identity MTS test was replicated, followed by a test of arbitrary MTS: trials involving letters and colors were presented, to test the emergency of conditional discrimination. Results suggested that 4 children developed conditional discriminations indicating the emergency of arbitrary stimulus classes. Results also indicated that although the discrimination training was effective in producing simple discrimination, it was not sufficient to guarantee good performance when identity-matching-to-sample trials were presented / No presente trabalho um procedimento de discriminação simples simultânea sem erro - dica atrasada - entre pares de letras graficamente semelhantes e suas inversões. Testou-se também os possíveis efeitos do treino discriminativo em tentativas de emparelhamento de acordo com o modelo (matching to sample) de identidade e arbitrário em que as letras que participaram do treino discriminativo eram estímulos comparação. Nove crianças (de 3 a 6 anos) com dificuldades nas discriminações entre pares de letras graficamente semelhantes e suas inversões (b-d, n-u, p-q e inversões em 270o e B-D, N-U, P-Q e inversões em 180 o) participaram da pesquisa. Depois de um pré-teste de MTS de identidade em que eram estímulos modelo letras e estímulos comparação as mesmas letras, letras semelhantes e suas inversões, as crianças selecionadas passaram por treinos discriminativos simples simultâneos, envolvendo letras (S+) e inversões (S-). No treino, cores foram estabelecidas como estímulos discriminativos (S+) ou como S-. A partir daí tais estímulos foram sobrepostos a novos estímulos e tal sobreposição foi atrasada a cada tentativa bem sucedida até que o participante escolhesse o sistematicamente o novo S+ (letra) antes da sobreposição de cores entre letras e suas entre cores (previamente após o treino, foi feito um pós- teste semelhante ao pré teste (MTS) de identidade) e um teste de matching arbitrário envolvendo as letras treinadas, suas inversões e as cores como estímulos modelo e comparação, para verificar a possível emergência de discriminações condicionais entre letras e cores, o que seria indicativo da formação de classes de estímulos equivalentes. Todos os participantes tiveram desempenhos bem sucedidos no treino discriminativo - quase sem erros. Quatro participantes tiveram desempenhos nos testes de MTS arbitrário que atestariam a emergência de discriminações condicionais que indicariam a possibilidade de formação de classes de estímulos. Os resultados dos pós-testes de MTS de identidade indicaram que para vários participantes o treino de discriminação simples não foi suficiente para garantir um bom desempenho na situação de MTS. Finalmente, em geral o desempenho dos participantes foi melhor quando os estímulos foram as letras maiúsculas