Tesi sul tema "Slos"

Despite the prevalence of student learning objectives (SLOs) in teacher evaluation systems throughout the United States, research on the validity of student and teacher SLO scores used for high-stakes decisions is lacking. For this reason, this dissertation is comprised of two chapters that examine student and teacher-level SLO performance data from select districts in one Race to the Top state. In Chapter 1, I describe the quality of student assessment data and the comparability of student scores across alternative growth targets. I find that in the first year of implementation, assessments from half of the courses in the sample contained indicators of poor data quality, including anomalous score distributions and small to negative correlations between student prescores and postscores. However, in the second year of implementation, when student SLO performance is incorporated into final teacher evaluation scores, far fewer assessments contained anomalous score distributions, and there is no evidence to suggest manipulation of student scores. In addition to the assessments, the choice of student growth target does have an impact on the comparability of student and teacher scores across districts and years. Chapter 2 describes the validity and reliability of teacher SLO scores. I find that while teacher SLO scores are moderately stable across courses, they are not stable over time, likely due to changes made to the assessments and targets used to determine student SLO scores. Further, for teachers with both SLO scores and an alternative metric of performance based on student growth, the two metrics do not converge. Finally, teachers in courses with higher average student prescores and lower proportions of students with disabilities have slightly higher SLO scores. In general, results on teacher SLO scores were similar to those found with value-added based metrics of teacher performance. Findings from both chapters suggest that improvement in the quality of the assessments administered as well as greater consistency in the growth targets assigned to students, both within districts over time and across districts, will improve the validity of student and teacher SLO scores in this state.
Quantitative Policy Analysis in Education

Large-conductance-Ca2+-activated potassium Slo1 channels are well characterised members of the Slo channel family. The other members, Slo3 and Slo2.χ, are less characterised and their physiological roles are elusive. Little is known of the pH-sensitive Slo3 channel, aside from a role regulating Ca2+ influx through CatSper channels during capacitation in spermatozoa. This thesis aims to investigate mSlo3 pharmacology and determine if mSlo3- EST’s encode novel isoforms. Na+-activated Slo2.χ channels (Slo2.1/Slo2.2) are thought to mediate adaptation of neuronal action potential firing rate and slow after hyperpolarisations following repetitive firing. We aim to characterise hSlo2.2a currents and their bithionol sensitivity, an activator of a rat isoform. Mutation of the conserved phenylalanine in the S6-domain (mSlo3-F304/hSlo1-F380) to tyrosine has similar effects on Slo channels, namely a negative shift in the voltagedependence of activation and an increase in membrane conductance. mSlo3 pharmacology matches the IKSper profile and suggests roles for mSlo3 as a pharmacological target in infertility treatment. This work identified a novel splice-variant of mSlo3; CV562866 lacks transmembrane domains and is more widely expressed than mSlo3. CV562866 protein is expressed in vivo, and contains the nominal RCK2 domain. CV562866 affects functional expression of mSlo3 and mSlo1 channels in Xenopus oocytes, decreasing and increasing expression respectively. This study examined the hSlo2.2a channel characteristics, which unlike the previously reported rSlo2.2a channel, mediates a non-rectifying K+ current. Bithionol activates hSlo2.2a channels with an EC50 of 1.29±0.12μM and activation was Na+ co-operative. We found a candidate bithionol binding site in the C-terminal domain near the “Cl- bowl”. Due to toxicity of hSlo2.1 expression; determination of the bithionol selectivity of Slo2.χ channels was unattainable. In vivo, Slo2.χ channels act as a background K+ conductance, activated by a variety of conditions, stabilising the resting membrane potential of the cell. This study suggests Slo2.χ are viable pharmacological targets for treatment of inflammatory pain.

L'augmentation de l'interaction lumière-matière aux échelles micro et nanométriques est un des fers de lance de la nanophotonique. En effet, le contrôle de la répartition spatiale de la lumière grâce à l'interaction résonante entre nanostructures et ondes électromagnétiques a conduit aux développements de nombreuses applications dans des domaines variés tels que les télécommunications,la spectroscopie et la détection d'objets. Le ralentissement de la lumière, sujet de la thèse, obtenue grâces à l'interférence d'ondes contre-propageantes dans des milieux périodiques ou le confinement sub-longueur d'onde dans des guides d'ondes plasmoniques, est associé à une compression des pulses lumineux et une forte augmentation du champ électrique, deux phénomènes clés pour la miniaturisation de composées optiques et l'augmentation de l'interaction lumière matière
Enhancing light-matter interactions at micro and nanoscales is one of the spearheads of nanophotonics. Indeed, the control of the field distribution due to the resonant interaction of nanostructures with electromagnetic waves has prompted the development of numerous optical components for many applications in telecommunication, spectroscopy or sensing. A promising approach lies in the control of light speed in nanostructures. Light slowdown, obtained by wave interferences in periodic structures or subwavelength confinement in plasmonic waveguides, is associated to pulse compressions and large field enhancements,which are envisioned as key processes for the miniaturization of optical devices and the enhancement of light-matter interactions.The thesis studies both fundamental aspects and possible applications related to slow light in photonic and plasmonic nanostructures. In particular, we study the impact of periodic system sizes on the group velocity reduction and propose a novelfamily of resonators that implement slow light on very small spatial scales. We then investigate the role of fabrication disorder in slow periodic waveguides on light localization and demonstrate how modal properties influence the confinement of localized modes. Also we propose a new hollow-core photonic crystal waveguide that provides efficient and remote couplings between the waveguide and atoms thatare trapped away from it. Finally we demonstrate the important role played by slow plasmons on the emission of quantum emitters placed in nanogap plasmonic antennas and explain how large radiation efficiency can be achieved by overcoming quenching in the metal. Additionally, one part of the thesis is devoted to thederivation of a novel modal method to accurately describe the dynamics of plasmonic resonators under short pulse illumination

Viele zeitgenössische Institutionen (Theater, Festivals, Biennalen, Kunsthallen, Museen, etc.) haben sich seit Mitte der 1990er Jahre wieder verstärkt experimentellen Formen zugewandt, sowie einen Theorie und Praxis verzahnten Diskurs gefördert. Wie aber sieht es in der Realität mit einer kontinuierlichen Programmarbeit in diesem Bereich aus? Gehören diese Formate und programmatischen Überschriften lediglich zu einer aufgeklärten Gesellschaft des Spektakels, die nach mehr Abwechslung verlangt oder existieren substantielle und kulturpolitische Langzeitperspektiven?

This thesis is a collection of original poems. The poems found herein attempt to bridge a perceived gap between poet and audience by using clear, interesting language and relatable subject matters. The setting for many of these poems is New England, and travel between the American Northeast and the South is emphasized. The speakers in these poems often find themselves positioned between two contrasting worlds and attempt to negotiate between them. Consistent themes include travel, nature, generation, family, and growing up.

Dissertação (mestrado) - Universidade Federal de Santa Catarina, Centro Sócio-Econômico, Programa de Pós-Graduação em Administração, Florianópolis, 2009
Made available in DSpace on 2012-10-24T07:50:46Z (GMT). No. of bitstreams: 1 269348.pdf: 879589 bytes, checksum: 5d5d1634df975dcb92c3eef779836ab4 (MD5)
Esta dissertação objetiva a percepção dos enclaves do ecodesenvolvimento no movimento slow. Para este diagnóstico, contextualizam-se as teorias tradicionais e os modelos de desenvolvimento pela racionalidade instrumental, que pressupõem o mercado como única arena das necessidades humanas, sociais e ambientais. Contudo, criticam-se as conseqüências negativas do nivelamento mercantil e sua unidimensão. Vários autores respondem que o desenvolvimento deve fomentar a racionalidade substantiva para uma mudança de paradigma em oposição ao maldesenvolvimento. O ecodesenvolvimento, enquanto planejamento, volta-se ao território para alavancar o desenvolvimento com respeito à diversidade ecossistêmica e à cultura local, o suporte de tecnologias para o uso de parcimonioso de recursos naturais e de tecnologias limpas. Também, percebe as relações de demanda pelos modelos culturais da sociedade, seu estilo de vida e suas estruturas de consumo como indicadores da legitimação de outro modelo de desenvolvimento. Os movimentos slow apontam a necessidade de uma desaceleração (quebra da lógica produtiva) e o resgate de valores integrativos e colaborativos pela sociedade, considerando várias esferas de necessidades humanas. Seus projetos têm forte ligação com o desenvolvimento territorial sustentado e com a preservação ambiental e cultural das localidades. Considera-se, então, o rumo de uma globalização virtuosa, em que se respeitem as pequenas produções e o comércio local com intuito da não massificação cultural e utilização responsável dos recursos.
This thesis aims at understanding the enclaves of ecodevelopment in slow movement. For this diagnosis, contextualized the traditional theories and models of development by instrumental rationality, which require the market as the only arena of human needs, social and environmental. However, criticize it the negative consequences of race and one-dimension market. Several authors reply that the development should encourage the substantive rationality for a change of paradigm in opposition to bad-development. The ecodevelopment as planning, returns to the territory to leverage the development with respect to ecosystem diversity and local culture, support technologies for the thrifty use of natural resources and clean technologies. Also, understands that the relationship of demand for society's cultural models, its lifestyle and its structures of consumption as indicators of legitimacy of another model of development. The slow movements suggest the needs for a deceleration (decline of productive logic) and the redemption of integrative value and collaborative by the society considering various spheres of human needs. Its projects have strong links with territorial sustainable development with environmental and cultural preservation of locations. It considers that the direction of a virtuous globalization, where is respect for small local production and trade with intention of no mass culture and responsible use of resources.

Research into how one can create architecture that will last beyond the foreseeable future. Part ONE is research based, building up a manifesto based on; structural integrity, aesthetic longevity, programmatic adaptability and cultural relevance. The research is based on observing historical architecture from the past 2,000 years, both looking into how the buildings have stood the test of time physically, but also why they might remain attractive, analysing the comings and goings of beauty through time. The research is compiled and represented through suggested ideal ways of designing each of the architectural elements, such as the wall, window, door, threshold or stair. Part TWO applies the research to a case study; the extension and renovation of the Royal Palace in Stockholm, in order to turn the residence into a public building, with the extension functioning as a connection point between the city landscape and the palace beyond. The design has no specific programme other than allowing the palace complex to be used by the public, as they wish, for the future to come. A large part of the design is essentially made up of sculptural urban landscaping, colonnaded and vaulted walkways both interacting with the waterfront, and bringing the public up to the level of the palace which today stands behind tall walls. Whilst internally the new building functions as an entrance hall with necessary facilities, and creates a route into the existing palace.

Like the ever-circling lines in the skin of trees, sometimes the whole of a person is peeled back, layer by layer, until only seeds remain. Names, faces, stories, and relationships are unmade and molded into new shapes. Without warning, those left at the base, at the roots, can’t recognize this maple’s form. They hold a pile of leaves, a bottle of glue, and the hope that something familiar will take sprout again. The tree becomes new, and its tangled branches reach out. These poems explore the lives of those living with and caring for those with dementia and Alzheimer’s disease. The hope is that they will touch on all the layers that make up a person, and the connection that nature and materiality have with building memory.

Ett av skolans viktigaste uppdrag är att lära eleverna att läsa och skriva, vårt samhälles kommunikationssystem litar på att alla har dessa färdigheter. (Lundberg, 2003) På skolorna finns det en mängd olika kartläggningsscheman som ska vara en hjälp för läraren att dokumentera elevernas läsutveckling så att skolsystemet inte missar elever som fastnar i utvecklingen. Ett nytt sådant kartläggningsschema som utvecklas i Älmhults kommun är SLUS, språk- och läsutvecklingsschema. Studiens syfte är dels att belysa utvecklingen av SLUS samt införandet av materialet i en rad kommuner dels att synliggöra hur verksamma pedagoger upplever arbetet med materialet SLUS. Jag använder mig av semistrukturerade intervjuer för att genomföra studien. SLUS är enbart ett kartläggningsschema med det finns ett material, God läsutveckling i praktiken, där det finns uppgifter som väl kan användas tillsammans med schemat för att bedöma barnets kunskaper. Tillsammans är det ett material som används aktivt i både Alvesta och Älmhults kommun och många pedagoger är mycket positiva till materialet.

Thesis (Master of Architecture)--University of Cincinnati, 2009.
Advisors: John Hancock (Committee Chair), Gerald Larson (Committee Co-Chair), Terry Boling (Advisor). Title from electronic thesis title page (viewed July 25, 2009). Includes abstract. Keywords: Slow food; regional; local; place. Includes bibliographical references.

Slow tourism has emerged as a counter reaction to increased competition and acceleration in today’s world and can be interpreted as both an emerging tourism niche as well as an overall approach to travel. While the literature on slow tourism is still evolving, Latvia with its present tourism marketing brand “Latvia. Best enjoyed slowly” is officially promoted as an appropriate destination for slow tourism and is currently the only destination that has applied the slow philosophy in destination marketing at a national level making the setting of the research unique. This study aimed to get an in-depth understanding of slow tourism, slow tourists and the application of slow philosophy in destination marketing and management. This was achieved by adopting a qualitative approach. Interviews with both the supply and demand side representatives were used to collect data and thematic analysis was employed to analyse them. This study makes the following original contributions. In terms of the theoretical contributions the findings confirm the previous concept of slow tourism as experience-based tourism but argue that slow tourism practices are influenced mainly by external factors especially in relation to mobilities. In particular, the findings show that the mindset of the tourist is the key criteria for identifying slow tourists. This study also suggests that a slow destination brand has the capacity to work not only as a pull factor but also as a push factor especially for local tourists. In addition to the push and pull factors already described in the literature, nature and patriotism were identified as important pull factors for slow tourism. In terms of the policy recommendations the thesis highlights the need to develop a practical toolkit for implementing a slow destination brand. The methodological contribution is to use specific criteria to identify people engaging in slow tourism rather than look for self-identified slow tourists in non-English speaking countries.

Because of environmental pollution, forces the automotive industry constantly reduced emissions requirements legislated by the authorities. Improved techniques for engine control are a must for bringing down emissions. The use of an Exhaust Gas Recirculation (EGR) reduces NOx emissions significantly. Faulty EGR valves affect the emissions negative and therefore needs to be eliminated. It is possible to create malfunctions on the EGR valve by modifying the software of the control unit (ECU), but it does not create realistic malfunctions. The problem by modifying the software is that flags and various parameters are set to confirm the malfunction of the ECU. To create actual failure of the EGR valve an external tool to modify the control signal is needed. The project's main objective is on a flexible way creating malfunctions on the EGR valve in a truck engine. By investigating engine behavior in a realistic and credible way, one can eliminate malfunctions on the EGR valve. The aim was achieved by a model that has been developed that can, using electronics and a microprocessor read and create a control signal. The electronic circuit is controlled by the microprocessor, which can modify the signal and create malfunctions in the form of a slow valve "slow-response". A graphical user interface is used to change and influence the error signal. The circuit with the microprocessor is placed safely in a box to both protect and preserve the components. Simulation of Slow response has resulted in an incorrect operated valve being created. Using two different methods a Slow-response can be created. One method is a delay in time, which occurs when the new position is given, the second method is a ramp function when the control signal is gradually increasing. The software can also create an error that mimics a stuck valve of a fixed value. With the above listed methods it is possible in theory to find unknown malfunctions on the EGR valve that influence emissions negatively.

Main purpose of the thesis is to study the processes and methods of isolated words recognition. In the theoretical part a basic principals are explained. The practical part is about the program creating using these principles in practice. For isolated words recognition Hidden Markov Models (HMM) are used, for obtaining decision symptoms cepstral analysis is chosen.

MARINHA DO BRASIL
Os sistemas que operam em banda larga necessitam de radiadores que atendam eletricamente a banda de freqüência utilizada. As antenas com características de independência da freqüência, em particular as antenas espirais, têm encontrando cada vez mais utilização nesses sistemas, seja em aplicações comercias ou militares. Esta dissertação tem por objetivo a análise de desempenho de antenas espirais do tipo fenda em cavidades, pela aplicação do método dos momentos. Serão consideradas cavidades com a face oposta à antena constituída por um condutor perfeito ou acoplada a uma estrutura de microondas com matriz de espalhamento conhecida. Essa estrutura de microondas, poderá simular, por exemplo, uma ou mais camadas de material absorvente.Um modelo numérico foi desenvolvido, e a partir dele foi elaborado um programa de computador, para determinar as características de radiação de antenas espirais em cavidade. São apresentados e comparados os resultados obtidos para três configurações de antenas espirais de Arquimedes do tipo fenda: sem cavidade, com cavidade condutora e com um material absorvente colocado sobre a face da cavidade oposta a abertura da antena.
Broad band systems need radiators to operate over the entire frequency band. Frequency independent antennas, in particular spiral antennas, are finding increasing application in such systems, for commercial or military purposes. This dissertation has for objective the analysis of performance of cavity-backed slot spiral antenna, by applying the moment method . For this study it will be taken into account cavities with the face opposing to the antenna consisting of a perfect conductor or connected to a structure of microwaves with known scattering matrix. This microwave structure will be able to simulate, for example, one or more layers of absorbing material. A numerical model was developed, and based on it, it was elaborated a computer program to determine the radiation characteristics of cavity-backed spiral antennas. The results surveyed for three configurations of slot Archimedean spiral antenna are presented and compared: without cavity, with conducting cavity, and with an absorbing material placed on the cavity opposite to the antenna.

This thesis is focused on the problem of semantic similarity of words in English language. At first reader is informed about theory of word sense clustering, then there are described chosen methods and tools related to the topic. In the practical part we design and implement system for determining semantic similarity using Word2Vec tool, particularly we focus on biomedical texts of MEDLINE database. At the end of the thesis we discuss reached results and give some ideas to improve the system.

This thesis surveys theoretical background for automatic keyword suggestion system. It contains overview of current statistical term recognition methods and methods for evaluation of automatic term recognition systems. Based on the known approach the thesis specifies possible enhancements. It explores unifying keywords using thesauri, input text filtering and correction of word forms.

Sweaty Mother Slow Groove is an engagement in magical thinking that proposes a displacement of swamp methodologies into the virtual realm, existing during the fourth wave. In doing so the cyborg and goddess are united in a re-routing of essentialism and the neo-liberal domination of technology. The metaphorical swamp is the possibility of a mushy danger zone that harnesses the absorption of an unwanted space: a disintegration of the binary and the soft-coded awareness of the body as a process, not a site.

This thesis is focused on lemmatizing of nouns and adjectives. It is based on morphology of Czech language. The goal is to create a lemmatizer which can stem words with success rate 90% (at least). At the same time the lemmatizer should be very easy, it should consist as little rules as possible. Lemmatizer will be created to work with real estate adverts, especially houses for sale. In this thesis there will be made an analysis of specific characters of this area. Lemmatizer will be created according to results of this analysis. Lemmatizer was written in Java. Only three types of rules were used and generally the lemmatizer created correct stems in 96.4% of all words.

Thesis (Ph. D.)--Joint Program in Oceanography (Massachusetts Institute of Technology, Dept. of Earth, Atmospheric and Planetary Sciences, and the Woods Hole Oceanographic Institution), 2001.
Page 250 blank.
Includes bibliographical references.
Half of the ocean crust is formed at spreading centers with total opening rates less than 40 km/Myr. The objective of this Thesis is to investigate temporal variations in active ridge processes and crustal aging at slow-spreading centers by comparing axial crustal structure with that on conjugate flanks of the slow-spreading Mid-Atlantic Ridge (MAR) (full rate, 20 km/Myr) and the ultra-slow spreading Southwest Indian Ridge (SWIR) (full rate, 14 km/Myr). Seismic refraction data collected along the rift valley and flanking rift mountains of the OH-1 segment (35ʻN) at the MAR show that the entire crustal section is constructed within a zone that is less than 5 km wide. Shallow-level hydrothermal circulation within the axial valley is suggested by the rift mountain seismic profiles, which show that the upper crust is 20% thinner and 16% faster along strike than zero-age crust. These effects probably result from fissure sealing within the extrusive crust. Deeper crustal velocities remain relatively constant at the segment midpoint within the first 2 Myr, but are reduced near the segment offsets presumably by faulting and fracturing associated with uplift out of the rift valley.
(cont.) A temporal variation in axial melt supply is suggested by a 15% difference in along-strike crustal thickness between the rift valley and rift mountains, with relatively less melt supplied today than 2 Ma. Crustal accretion at the SWIR appears to occur in a similar manner as at the MAR, although gravity and seismic data indicate that the average crustal thickness is 2-4 km less at theultra-slow spreading SWIR. A 25 Myr record on both flanks of the ridge shows that seafloor spreading has been highly asymmetric through time, with 35% faster crustal accretion on the Antarctic (south) plate. A small-offset non-transform discontinuity between two ridge segments is just as stable as two neighboring transform discontinuities, although a single mantle Bouguer gravity anomaly centered over the non-transform offset indicates that this boundary does not significantly perturb underlying mantle flow. Off-axis magnetic anomalies are recorded with high fidelity despite the very low spreading rates and the absence of a basaltic upper crust in one area. The lower crust may be the dominant off-axis carrier of the magnetic signal, contrary to traditionalmodels of crustal magnetic structure. Morphological and gravity data show evidence of asymmetric crustal accretion across the SWIR ridge axis, with slightly warmer mantletemperatures beneath the slower-spreading African (north) plate.
by Allegra Hosford.
Ph.D.

International air transport has grown at a staggering rate over the last 50 years and the industry which supports it, now bears little resemblance to that which existed even twenty-five years ago. As the number of operators and the size of their respective fleets have increased, so too have the pressures on the infrastructure which support them. Particularly affected are the major airports where runway, apron and terminal capacity are limited. Unlike other elements of the aviation infrastructure, airport capacity is physically constrained and therefore capacity increases can only be obtained through further development or through more efficient uses of existing resources. As the first option is politically controversial and the latter only provides for incremental increases, the air transport industry, through IATA, has developed an allocation mechanism to distribute available capacity based on historic precedence. This thesis reviews the main issues surrounding the current regime. (Abstract shortened by UMI.)

Thesis (M.S. in Ocean Engineering) Woods Hole Oceanographic Institution and the Massachusetts Institute of Technology, Sept. 1997.
DTIC Descriptor(s): Underwater Vehicles, Autonomous Navigation, Optimization, Adaptive Control Systems, Pitch (Motion), Equations Of Motion, Hydrodynamic Characteristics, Performance (Engineering), Theses, Low Velocity, Control Theory, Energy Conservation, Submarine Models. Includes bibliographical references (leaves 101-104). Also available online.

This thesis describes a novel experimental system for the study of polymer flow in a cross-slot using a Cambridge Multi-Pass Rheometer (MPR). Cross-slot flow is capable of generating pure rotation-free extensional flow and the rheological information from the MPR is in a format which can easily be compared with the results of numerical simulations. Using only a small quantity of polymer, this technique provides a quantitative test of the performance of constitutive equations and numerical solvers in extensional flow. The cross-slot apparatus was developed and experiments were studied principally using optical birefringence. It was found that the different shapes of the birefringence patterns observed reveal useful information about the rheology and molecular structure of the polymers. For unbranched low molecular weight polymers at low extension rates, an almost Newtonian response was observed, generating a symmetric fringe pattern. as the level of branching, the molecular weight or the extension rate increased, the fringe pattern became more elongated, suggesting stronger viscoelastic effects. For highly branched polymers a very localised stress concentration was observed along the exit symmetry plane. Pioneering experiments with monodisperse polymer have been particularly revealing – a flow transition is observed at extension rates approximately equal to the inverse of the relaxation time. At low flowrates the response was almost Newtonian, but for extension rates above the inverse of the relaxation time a non-homogeneous stress pattern was observed suggesting possible melt fracture of the fluid in the continuum. The experiments have been compared with numerical simulations using generalised Newtonian, integral-Wagner, Rolie-Poly and pom-pom constructive equations. It has been found that the viscoelastic numerical simulations revealed many of the important features observed from the experiments. However, the quantitative value for stress predicted in some case was higher than the measured value.

PONTIFÍCIA UNIVERSIDADE CATÓLICA DO RIO DE JANEIRO
COORDENAÇÃO DE APERFEIÇOAMENTO DO PESSOAL DE ENSINO SUPERIOR
FUNDAÇÃO DE APOIO À PESQUISA DO ESTADO DO RIO DE JANEIRO
PROGRAMA DE EXCELENCIA ACADEMICA
O processo de revestimento por extrusão é muito usado na fabricação de produtos de alta tecnologia tais como circuitos flexíveis, telas LCD e OLED, dentre outros. A espessura do filme revestido depende da vazão de alimentação e da velocidade do substrato. O estudo do escoamento na região de aplicação é de muita importância, devido à influência que tem sobre a qualidade do filme revestido. A grande maioria dos trabalhos disponíveis na literatura consideram o líquido como um fluido Newtoniano. Porém, nos processos de revestimentos nas indústrias, os líquidos comumente usados são soluções poliméricas e suspensões de partículas. O efeito da presença de moléculas poliméricas dissolvidas no fluido de trabalho nos padrões de escoamento e consequentemente na janela de operação do processo foi estudado recentemente, contudo, o efeito de partículas suspensas no escoamento é ainda desconhecido. O objetivo deste trabalho é compreender melhor a influência da suspensão de partículas no padrão de escoamento que ocorre na região de aplicação de um filme fino de líquido sobre um substrato em movimento. A variação local da concentração de partículas no líquido pode levar a grande variações na viscosidade e consequentemente mudanças no escoamento e nos limites do processo. Neste estudo, considera-se a suspensão como um sistema formado por um fluido Newtoniano como a fase contínua e partículas esféricas rígidas como a fase dispersa. A equação de transporte de partículas engloba os efeitos de difusão de partículas devido a gradientes de concentração, viscosidade e taxa de deformação. As equações de transporte de partículas em conjunto com as equações de conservação de massa e quantidade de movimento são usadas para descrever o escoamento bidimensional com fronteiras livres. Estas equações são resolvidas aproximadamente pelo método de Galerkin/Elementos Finitos. A validação do código numérico foi feita comparando com resultados analíticos disponíveis para escoamento em tubos de seção circular. Os resultados obtidos para o escoamento no precesso de revestimento por extrusão mostram as regiões de maior e menor concentração de partículas e como esta variação altera o escoamento, e a variação da concentração ao longo da espessura do filme depositado.
Slot coating process is commonly used in the manufacture of high technology products such as flexible circuits, LCD and OLED displays, among others. The thickness of the coated liquid film depends on the flow rate and substrate velocity. It is very important to study the flow in the coating bead region, because the flow pattern has a strong effect on the quality of the coated film. Most of the available studies on slot coating consider the fluid as a Newtonian liquid. However, coating liquids are usually polymer solutions or particle suspensions. The effect of dissolved polymer molecules in the flow has been analyzed recently. However, the effect of suspended particle remains unknown. The goal of this work is to develop a fundamental understanding of the effect of suspended particles in the flow pattern in a slot coating process. The local changes in particle concentration can lead to great changes in the local liquid viscosity and consequently in the flow and process limits. In this research, the particle suspension is a system composed by a Newtonian liquid as a continuous phase and rigid spherical particles as the dispersed phase. The particle transport equations takes into account the diffusion due to gradients of concentration, viscosity and deformation rates. The particle transport equation together with the mass and momentum conservation equations are used to describe the two-dimensional free surface flow. These equations are solved using the Galerkin/Finite element method. The numerical code was validated by comparing numerical prediction obtained in circular tube flow with the available exact solution. The results obtained in slot coating flow show regions of high and low concentration of particles and its effect of the flow, and the variation of particle concentration in the deposited thin film.

CONSELHO NACIONAL DE DESENVOLVIMENTO CIENTÍFICO E TECNOLÓGICO
O processo de revestimento por extrusão é um método muito utilizado na manufatura de diversos produtos. Ele pertence à uma classe de processos de revestimento chamada revestimento com vazão pré-fixada: para operações em regime permanente, a espessura da camada de líquido revestida é definida pela vazão na entrada do equipamento e pela velocidade do substrato. Para diversas aplicações, o líquido de revestimento é uma suspensão de partículas. A abordagem comumente usada é analisar o escoamento como se o fluido fosse Newtoniano, onde sua viscosidade é avaliada pela concentração média de partículas. Porém, dados experimentais mostram que a distribuição de partículas não é uniforme em escoamentos com cisalhamento. Além disso, a distribuição de partículas no filme após o processo de revestimento afetará a estrutura final do filme e, consequentemente, a qualidade final do produto. Portanto, é importante entender os fundamentos do processo de revestimento com suspensão de partículas para melhor descrever este processo e prever comportamentos desse escoamento. O objetivo do presente trabalho é estudar diferentes mecanismos de migração de partículas em suspensão que podem afetar a distribuição de partículas no filme revestido; como mecanismos de difusão e sedimentação. O modelo apresentado leva em consideração a dependência da viscosidade pela concentração local de partículas e o efeito de Marangoni. Dois modelos diferentes foram usados para estudar o escoamento de Marangoni. No primeiro, o transporte de partículas entre a superfície e o escoamento é considerado muito mais rápido que a difusão no escoamento de tal forma que o fluxo líquido é zero e a concentração de partículas na região do escoamento próxima à superfície. No segundo, um modelo mais completo, os efeitos de adsorção e dessorção de partículas pela interface é considerado. O sistema final de equações e não-linear e, com as condições de contorno apropriadas, é resolvido pelo método de Galerkin em elementos finitos e pelo método de Newton. O escoamento é bi-dimensional e, portanto, elementos bi-dimensionais são utilizados para descrever o domínio do problema. Entretanto, para avaliar a concentração ao longo da superfície, foi necessário criar um elemento uni-dimensional. O acoplamento entre estas duas classes de elementos também é discutido. Os resultados mostram que a distribuição de partículas do filme revestido é uma forte função da espessura do filme e das propriedades da suspensão, como a densidade das partículas e coeficientes de difusão. O escoamento de Marangoni pode afetar o padrão de recirculação e a distribuição de concentração de partículas. Além disso, os resultados mostram que a adsorção e dessorção de partículas na interface afetam fortemente a distribuição de partículas.
Slot coating is a common method in the manufacturing of a wide variety of products. It belongs to a class of coating method known as premetered coating: in a steady state operation, the thickness of the coated liquid layer is set by the flow rate fed to the die and the speed of the substrate moving past, and is independent of other variables of the process. For many applications, the coating liquid is a particle suspension. The commom simplified approach is to study the flow as Newtonian and evaluate its viscosity based on the average particle concentration. However, experimental data shows that particle distribution is in fact non-uniform in shear flows. Moreover, particle distribution along the film thickness during the coating process will affect the final fim structure and consequently product performance. Hence, it is important to understand the fundamentals of coating process with particle suspension to better describe and predict the behavior of the flow and the particle distribution in the coated layer. The goal of the present work is to study different particle migration mechanisms in a suspension that may affect particle distribution in the coated film; such as diffusive mechanisms and sedimentation. The model presented takes into account the viscosity dependency and local particle concentration and surface-active particles, i.e. Marangoni effect. Two different approaches were used to study Marangoni flow. First, the bulk-interface transport is considered much faster than bulk diffusion such that the net flux is zero and surface concentration is equal bulk concentration. The second, more complete model, considers adsorption and desorption fluxes from the interface. The non-linear system of equations, with appropriate boundary conditions, is solved by Galerkin Finite Element Method and Newton s Method. The flow is two-dimensional and therefore two-dimensional elements are used to describe flow domain. Nonetheless, in order to evaluate particle concentration along the interface a one-dimensional element was created. The coupling between those two classes of elements of discussed. The results show that the particle distribution on the coated layer is a strong function of the film thickness and suspension properties, such as particle density and diffusion coefficients. The Marangoni flow associated with surface tension gradient due to particle concentration variation along the interface can change the recirculation pattern and particle concentration distribution. Furthermore, the results show that particle adsorption and desorption from interface have a strong effect on the particle distribution.

L'ère du Big Data a révolutionné la manière dont les données sont créées et traitées. Dans ce contexte, de nombreux défis se posent, compte tenu de la quantité énorme de données disponibles qui doivent être efficacement gérées et traitées afin d’extraire des connaissances. Cette thèse explore la symbiose de l'apprentissage en mode batch et en flux, traditionnellement considérés dans la littérature comme antagonistes, sur le problème de la classification à partir de flux de données en évolution. L'apprentissage en mode batch est une approche bien établie basée sur une séquence finie: d'abord les données sont collectées, puis les modèles prédictifs sont créés, finalement le modèle est appliqué. Par contre, l’apprentissage par flux considère les données comme infinies, rendant le problème d’apprentissage comme une tâche continue (sans fin). De plus, les flux de données peuvent évoluer dans le temps, ce qui signifie que la relation entre les caractéristiques et la réponse correspondante peut changer. Nous proposons un cadre systématique pour prévoir le surendettement, un problème du monde réel ayant des implications importantes dans la société moderne. Les deux versions du mécanisme d'alerte précoce (batch et flux) surpassent les performances de base de la solution mise en œuvre par le Groupe BPCE, la deuxième institution bancaire en France. De plus, nous introduisons une méthode d'imputation évolutive basée sur un modèle pour les données manquantes dans la classification. Cette méthode présente le problème d'imputation sous la forme d'un ensemble de tâches de classification / régression résolues progressivement.Nous présentons un cadre unifié qui sert de plate-forme d'apprentissage commune où les méthodes de traitement par batch et par flux peuvent interagir de manière positive. Nous montrons que les méthodes batch peuvent être efficacement formées sur le réglage du flux dans des conditions spécifiques. Nous proposons également une adaptation de l'Extreme Gradient Boosting algorithme aux flux de données en évolution. La méthode adaptative proposée génère et met à jour l'ensemble de manière incrémentielle à l'aide de mini-lots de données. Enfin, nous présentons scikit-multiflow, un framework open source en Python qui comble le vide en Python pour une plate-forme de développement/recherche pour l'apprentissage à partir de flux de données en évolution
The Big Data era has revolutionized the way in which data is created and processed. In this context, multiple challenges arise given the massive amount of data that needs to be efficiently handled and processed in order to extract knowledge. This thesis explores the symbiosis of batch and stream learning, which are traditionally considered in the literature as antagonists. We focus on the problem of classification from evolving data streams.Batch learning is a well-established approach in machine learning based on a finite sequence: first data is collected, then predictive models are created, then the model is applied. On the other hand, stream learning considers data as infinite, rendering the learning problem as a continuous (never-ending) task. Furthermore, data streams can evolve over time, meaning that the relationship between features and the corresponding response (class in classification) can change.We propose a systematic framework to predict over-indebtedness, a real-world problem with significant implications in modern society. The two versions of the early warning mechanism (batch and stream) outperform the baseline performance of the solution implemented by the Groupe BPCE, the second largest banking institution in France. Additionally, we introduce a scalable model-based imputation method for missing data in classification. This method casts the imputation problem as a set of classification/regression tasks which are solved incrementally.We present a unified framework that serves as a common learning platform where batch and stream methods can positively interact. We show that batch methods can be efficiently trained on the stream setting under specific conditions. The proposed hybrid solution works under the positive interactions between batch and stream methods. We also propose an adaptation of the Extreme Gradient Boosting (XGBoost) algorithm for evolving data streams. The proposed adaptive method generates and updates the ensemble incrementally using mini-batches of data. Finally, we introduce scikit-multiflow, an open source framework in Python that fills the gap in Python for a development/research platform for learning from evolving data streams

This diploma thesis is focused on the detection and protection against Slow DoS and DDoS attacks using computer network traffic analysis. The reader is introduced to the basic issues of this specific category of sophisticated attacks, and the characteristics of several specific attacks are clarified. There is also a set of methods for detecting and protecting against these attacks. The proposed methods are used to implement custom intrusion prevention system that is deployed on the border filtering server of computer network in order to protect Web servers against attacks from the Internet. Then created system is tested in the laboratory network. Presented results of the testing show that the system is able to detect attacks Slow GET, Slow POST, Slow Read and Apache Range Header and then protect Web servers from affecting provided services.

This thesis pursues an automated extraction of keywords from documents. Its goal is to design and implement an application which will be able to extract an appropriate set of keywords related to the contents of the document. The major requirements for the application are speed and accuracy. That is why the first part of the thesis talks about already developed principles and a detailed classification based on various criteria. The second part is focused on choosing and a thorough functional describing of one of the methods which should have been used for extracting the keywords. The next parts contain a detailed draft of the application and its implementation. Finally, the last chapter is particularly important due to testing the application on a group of text documents and evaluating final results of the extraction process.

Psychologists have documented a panoply of beliefs that are sufficiently skewed towards desirability to arouse our suspicion that people believe things in part because they want them to be true (e.g. “above-average” effects (Alicke & Govorun, 2005; Baker & Emery, 1993; Beer & Hughes, 2010; Dunning, Meyerowitz, & Holzberg, 1989; Svenson, 1981; Williams & Gilovich, 2008), unrealistic optimism (Carver, Scheier, & Segerstrom, 2010; Scheier, Carver, & Bridges, 1994; Sharot, Korn, & Dolan, 2011; Weinstein, 1980), and wishful thinking (Aue, Nusbaum, & Cacioppo, 2011; Babad, 1997; Krizan & Windschitl, 2009; Windschitl, Scherer, Smith, & Rose, 2013)). The ostensible irrationality of these motivated biases poses a deep psychological question: how are such biases generated and maintained by a cognitive system that is presumably designed to accurately track reality? Studies that look at the motivated biases and the biased belief updating that may give rise to them tend to employ rich meaningful stimuli covering different targets of belief that are of every day concern: from your health, intelligence, and attractiveness, to your perfidy, academic performance, marital prognosis and driving ability. The use of such stimuli makes it difficult to account for the prior experience and beliefs relevant to such stimuli that a participant brings to the study as well as inadvertently reinforcing a view that motivated biases emerge through rumination upon specific and relatively sophisticated belief content (Lieberman, Ochsner, Gilbert, & Schacter, 2001). In this dissertation we changed this methodological emphasis. Over the course of the first three experiments, we demonstrate wishful thinking in a semantically sparse, repeated decision-making task about which participants can have no prior expectations, where the components of the task have no personal relevance beyond the experiment, and where they will be required to update their belief about the current state of affairs based upon a repeated and varying diet of desirable and undesirable evidence. We then situated this bias in the dual-process framework of judgment and decision-making by manipulating the time participants take to make their judgment in our task (Experiments 4a and 4b), by manipulating participants' cognitive load (Experiment 5), and by manipulating participants' thinking style—the weight participants put on the contribution from each type of processing—with an essay writing prime (Experiments 6a and 6b). On the whole, the results show that automatic processes alone are sufficient for wishful thinking. Though controlled, Type 2 processing inhibits the bias when induced to play a role, it does not typically contribute to the bias, either antagonistically or complementarily, absent such an inducement. Far from being an occasional, effortful rationalization that thrives on evidential complexity and uncertain costs, the wishful thinking bias we engendered is a simple, biased, belief updating process that operates automatically and beneath our awareness.
Psychology

Die Proteinfaltung ist ein Prozess der molekularen Selbstorganisation, bei dem sich eine lineare Kette von Aminosäuren zu einer definierten, funktionellen dreidimensionalen Struktur zusammensetzt. Der Prozess der Faltung ist ein thermisch getriebener diffusiver Prozess durch eine Gibbs-Energie-Landschaft im Konformationsraum für die Struktur der minimalen Energie. Während dieses Prozesses zeigt die freie Enthalpie des Systems nicht immer eine monotone Abnahme; stattdessen führt eine suboptimale Kompensation der Enthalpie- und der Entropieänderung während jedes Faltungsschrittes zur Bildung von Freien-Enthalpie-Faltungsbarrieren. Diese Barrieren und damit verbundenen hochenergetischen Übergangszustände, die wichtige Informationen über Mechanismen der Proteinfaltung enthalten, sind jedoch kinetisch unzugänglich. Um den Prozess der Barrierebildung und die strukturellen Merkmale von Übergangszuständen aufzudecken, werden Proteine genutzt, die über barrierefreie Pfade falten – so genannte “downhill folder“. Aufgrund der geringen Faltungsbarrieren werden wichtige Interaktionen der Faltung zugänglich und erlauben Einblicke in die ratenbegrenzenden Faltungsvorgänge. In dieser Arbeit vergleichen wir die Faltungsdynamiken von drei verschiedenen Varianten eines Lambda-Repressor-Fragments, bestehend aus den Aminosäuren 6 bis 85: ein Zwei-Zustands-Falter λWT (Y22W) und zwei downhill-folder-artige Varianten, λYA (Y22W/Q33Y/ G46,48A) und λHA (Y22W/Q33H/G46,48A). Um auf die Kinetik und die strukturelle Dynamik zu greifen zu können, werden Einzelmolekülkraftspektroskopische Experimente mit optische Pinzetten mit Submillisekunden- und Nanometer-Auflösung verwendet. Ich fand, dass die niedrige denaturierende Kraft die Mikrosekunden Faltungskinetik von downhill foldern auf eine Millisekunden-Zeitskala verlangsamt, sodass das System für Einzelmolekülstudien gut zugänglich ist. Interessanterweise zeigten sich unter Krafteinwirkung die downhill-folder-artigen Varianten des Lambda-Repressors als kooperative Zwei-Zustands-Falter mit deutlich unterschiedlicher Faltungskinetik und Kraftabhängigkeit. Drei Varianten des Proteins zeigten ein hoch konformes Verhalten unter Last. Die modellfreie Rekonstruktion von Freien-Enthalpie-Landschaften ermöglichte es uns, die feinen Details der Transformation des Zwei-Zustands-Faltungspfad direkt in einen downhill-artigen Pfad aufzulösen. Die Auswirkungen von einzelnen Mutationen auf die Proteinstabilität, Bildung der Übergangszustände und die konformationelle Heterogenität der Faltungs- und Entfaltungszustände konnten beobachtet werden. Interessanterweise zeigen unsere Ergebnisse, dass sich die untersuchten Varianten trotz der ultraschnellen Faltungszeit im Bereich von 2 μs in einem kooperativen Prozess über verbleibende Energiebarrieren falten und entfalten, was darauf hindeutet, dass wesentlich schnellere Faltungsraten notwendig sind um ein downhill Limit vollständig zu erreichen
Protein folding is a process of molecular self-assembly in which a linear chain of amino acids assembles into a defined, functional three-dimensional structure. The process of folding is a thermally driven diffusive search on a free-energy landscape in the conformational space for the minimal-energy structure. During that process, the free energy of the system does not always show a monotonic decrease; instead, sub-optimal compensation of enthalpy and entropy change during each folding step leads to formation of folding free-energy barriers. However, these barriers, and associated high-energy transition states, that contain key information about mechanisms of protein folding, are kinetically inaccessible. To reveal the barrier-formation process and structural characteristics of transition states, proteins are employed that fold via barrierless paths – so-called downhill folders. Due to the low folding barriers, the key folding interactions become accessible, yielding insights about the rate-limiting folding events. Here, I compared the folding dynamics of three different variants of a lambda repressor fragment, containing amino acids 6 to 85: a two-state folder λWT (Y22W) and two downhill-like folding variants, λYA (Y22W/Q33Y/G46,48A) and λHA (Y22W/Q33H/G46,48A). To access the kinetics and structural dynamics, single-molecule optical tweezers with submillisecond and nanometer resolution are used. I found that force perturbation slowed down the microsecond kinetics of downhill folders to a millisecond time-scale, making it accessible to single-molecule studies. Interestingly, under load, the downhill-like variants of lambda repressor appeared as cooperative two-state folders with significantly different folding kinetics and force dependence. The three protein variants displayed a highly compliant behaviour under load. Model-free reconstruction of free-energy landscapes allowed us to directly resolve the fine details of the transformation of the two-state folding path into a downhill-like path. The effect of single mutations on protein stability, transition state formation and conformational heterogeneity of folding and unfolding states was observed. Noteworthy, our results demonstrate, that despite the ultrafast folding time in a range of 2 µs, the studied variants fold and unfold in a cooperative process via residual barriers, suggesting that much faster folding rate constants are required to reach the full-downhill limit

Dissertação de Mestrado em Design, com a especialização em Design de Moda apresentada na Faculdade de Arquitetura da Universidade de Lisboa para obtenção do grau de Mestre.
A presente dissertação, de cariz teórico-prático, insere-se no âmbito do Design de Moda, particularmente na observação dos novos comportamentos do mercado e das novas tendências de consumo, resultantes do atual processo de globalização decorrente da Terceira Revolução Industrial. A partir de revisão de literatura relevante, este estudo pretende analisar a evolução da globalização e o seu impacto junto dos consumidores. Os antagónicos movimentos de Fast Fashion e Slow Fashion, assumem particular importância no contexto da indústria da Moda, e exprimem diferentes valores estéticos, culturais e sociais que proporcionam diferentes visões no pensamento e vivência dos consumidores, estabelecendo conexões entre as consequências geradas por ambos os movimentos. Através de uma análise SWOT, a presente investigação procura obter uma mais correcta identificação dos valores de cada um destes dois sistemas, fundamentais para a compreensão de uma moda mais consciencializada e justa. Com base nos princípios da Slow Fashion, que valorizam não só as questões da ética e sustentabilidade mas, e sobretudo, a valorização das habilidades tácteis dos artesãos populares portugueses, surge o levantamento das principais técnicas e desenhos utilizados na elaboração do tradicional “Bordado de Viana”, que servirá de suporte à coleção desenvolvida.
ABSTRACT: This essay, of a theoretical and practical nature, is aimed at the observation of new behaviors in markets as well as the new trends of consumption, resulting from the ongoing process of globalization of the Third Industrial Revolution. Starting with the consulting of relevant literature, this research intends to analysis the evolution of the globalization, and the impact on consumers. The antagonistic movements of Fast Fashion and Slow Fashion which are particularly important in the Fashion industry, and express different aesthetical, cultural, and social values that allow us new insights into the thinking and background of consumers, establishing connections between the consequences that arose from these movements. Through a SWOT analysis, the present investigation attempt to obtain a more accurate identification of the values of each of the two systems, which is fundamental in the understanding of a more aware and fair fashion. Based on the principles of Slow Fashion, which hold in high regard not only matters of ethics and sustainability but, first and foremost, the valuing of the labor skills of popular Portuguese artisans, comes up the main techniques and designs used in the production of the traditional “Bordado de Viana” were studied, and served as inspiration for the collection that will be developed.
N/A

Dissertação de Mestrado em Design, com a especialização em Design de Produto apresentada na Faculdade de Arquitetura da Universidade de Lisboa para obtenção do grau de Mestre.
prático e ancestralidade. A Arte Pastoril é uma arte milenar de produção de objetos utilitários que colmatam, de um modo engenhoso, as falhas que a carteira ou a vida não podiam fazer chegar. Esta Arte, no Alentejo Central, demonstra uma maior atenção ao detalhe, sendo praticamente irrepreensível em termos ergonómicos, funcionais e estéticos. Atualmente, a extinção progressiva desta Arte é já uma realidade pelo que, na presente investigação, se objetiva contribuir para o reconhecimento e valorização da genialidade dos artefactos da Arte Pastoril e contribuir, complementarmente, para o reposicionamento da Arte Pastoril do Alentejo Central. Utilizando os aspetos culturais como ponto de partida para a inovação pelo design, baseou-se a investigação numa metodologia mista de base qualitativa. Para a consolidação da componente teórica, e para a compreensão do contexto, recorreu-se à crítica da literatura e à identificação de casos de referência de parcerias entre design e artesanato, em Portugal. Posteriormente, na componente de investigação ativa, seguiu-se a participação num grupo de eventos considerados relevantes. Complementarmente, e no sentido de validar de forma preliminar o argumento, também foi realizado um conjunto de entrevistas e conversas informais. A investigação materializou-se na construção de uma ferramenta taxonómica, cuja validação se efetuou através de um grupo de foco. Na aplicação da ferramenta ao projeto exploratório, foi identificada a oportunidade de explorar a reinterpretação da colher provadeira, oriunda da Arte Pastoril alentejana, cujo potencial de reposicionamento foi identificado nas cozinhas de autor. A validação dessa hipótese fez-se junto de 13 de chefes de cozinha, cujas respostas foram amplamente favoráveis. Com a presente investigação, pretende-se, por um lado, motivar os designers de produto a olharem criticamente sobre este universo da Arte Pastoril e a contribuírem para a preservação do trabalho do artesão tradicional; por outro, possibilitar a aplicação desta ferramenta a outras áreas tradicionais em que se verifiquem as mesmas necessidades.
ABSTRACT: practical sense and ancestry. Pastoral Art is a millenary art of producing utility objects that fill, in an ingenious way, the flaws that the wallet or life could not make reach. This Art, in Alentejo Central, shows greater attention to detail, being practically blameless in ergonomic, functional and aesthetic terms. Currently, the progressive extinction of this Art is already a reality so that, in the present investigation, it aims to contribute to the recognition and valorization of the genius of pastoral art artifacts and contribute, in addition, to the repositioning of pastoral art of Central Alentejo. Using cultural aspects as a starting point for innovation by design, research was based on a qualitative-based mixed methodology. For the consolidation of the theoretical component, and for the understanding of the context, literature criticism and the identification of reference cases of partnerships between design and crafts in Portugal were used. Subsequently, in the active investigation component, participation in a group of events considered relevant followed. In addition, and in order to validate the argument in a preliminary way, a set of informal interviews and conversations was also conducted. The investigation materialized in the construction of a taxonomic tool, whose validation was carried out through a focus group. In the application of the tool to the exploratory project, the opportunity to explore the reinterpretation of the taste spoon, from the Alentejo Pastoral Art, whose potential for repositioning was identified in the author’s kitchens, was identified. The validation of this hypothesis was made with 13 chefs, whose answers were largely favorable. With this research, it is intended, on the one hand, to motivate product designers to look critically at this universe of Pastoral Art and to contribute to the preservation of the work of the traditional craftsman; on the other hand, enable the application of this tool to other traditional areas where the same needs are met.
N/A

L'apparition de nouvelles technologies comme l'informatique en nuages (cloud computing) offre de nouvelles opportunités de traitement de l'information. Par exemple, il est désormais facile de stocker ses photos ou vidéos personnelles sur des serveurs distants. Il est également possible de partager ces contenus à travers ces mêmes serveurs, ou encore via les réseaux sociaux ou les plateformes de téléchargement. Cependant, ces données personnelles sont bien souvent accessibles par le fournisseur de service, essentiellement pour des raisons pratiques : par exemple adapter une vidéo pour qu'elle s'affiche au bon format quel que soit l'appareil utilisé pour la visionner, permettre le partage de ses contenus avec d’autres personnes, etc. Cela soulève cependant un problème de confidentialité de ces données personnelles, et de confiance dans le fournisseur du service. La cryptographie classique apporte des solutions à ce problème, mais soulève malheureusement celui de la maniabilité des données : il devient par exemple impossible d'adapter un contenu vidéo au bon format d'affichage puisque le fournisseur ne peut plus « voir » la vidéo. Une solution alternative réside toutefois dans le chiffrement homomorphe. Cet outil un peu magique de la cryptographie avancée apporte la même sécurité que les algorithmes de cryptographie classique, mais permet de plus de manipuler les données tout en conservant leur forme chiffrée. Il offre ainsi une nouvelle perspective pour les fournisseurs puisque ceux-ci peuvent continuer à traiter l'information sans être capable de la voir, et donc sans atteinte à la vie privée de leurs utilisateurs, se conformant ainsi au nouveau Règlement Général sur la Protection des Données (RGPD). Bien que le chiffrement homomorphe soit très souvent considéré comme insuffisamment mature, du fait de sa complexité algorithmique, cette thèse cherche à montrer son caractère prometteur, en s'intéressant à son usage pour le traitement d'images et de vidéos chiffrées à la source. Nous regardons ainsi les différents algorithmes qui constituent un encodeur d'image/vidéo (JPEG/H264 et HEVC) et les transformons en des circuits qui sont manipulables par des systèmes de chiffrement homomorphes. Nous proposons ainsi dans cette thèse le tout premier pipeline de compression d'images de type JPEG ("homomorphic-JPEG") sur des pixels qui sont chiffrés de bout-en-bout. Pour optimiser la gestion des données ainsi protégées, nous proposons également de nouveaux outils applicables à tous les schémas de chiffrement homomorphe sur les réseaux idéaux. Notre approche permet de maximiser le nombre de slots dans un chiffré et introduit de nouvelles fonctions pour manipuler ces différents slots de manière indépendante les uns des autres. Ces travaux de thèse ont abouti à la publication de deux articles dans des conférences internationales ainsi qu’à la soumission d'un article supplémentaire
The emergence of new technologies like cloud computing brings new opportunities in information processing. For example it is easy today to send our personal pictures or videos to a remote server (Google Drive, OneDrive …). We can also share this content among the same servers or via social networks and streaming services. However, this personal data is often also available to the service provider, mainly for practical reasons e.g. to configure a video to have the right format regardless of the displayer (smartphone or computer), to share our data with other people, etc. This raises issues of privacy and trust into the service provider. Classical cryptography brings some answers to this kind of issues, yet leaving the problem of handling the encrypted data: e.g., it becomes impossible to reconfigure a video because the provider can no longer “see” it. An alternative solution is “homomorphic encryption”. It is a powerful tool of advanced cryptography which provides the same security as classical cryptography algorithms, but it still allows us to manipulate ciphertexts such their underlying plaintexts are modified. Consequently, it offers a new perspective to service providers since they can continue to process their clients’information without knowing what it contains. This allows them to provide privacy-preserving services and comply with the new General Data Protection Regulation (GDPR). Although it is considered that homomorphic encryption does not have enough maturity due to its large algorithmic complexity, in this thesis, we are trying to show its potential by using it in the context of image and video processing over the encrypted data. In this context, we look at the different algorithms in an image/video encoder (JPEG/H264 and HEVC) and transform them to circuits which can be manipulated by homomorphic encryption schemes. Our main contribution is to propose the first pipeline for an image compression of type JPEG (homomorphic-JPEG) running on end-to-end encrypted pixels. To optimize the management of the encrypted data, we also propose new tools applicable to existing homomorphic encryption schemes over the ring version of lattices. Our approach allows us to maximize the number of slots in some ciphertext and we introduce new functions allowing to handle these slots independently in the encrypted domain. This thesis work also lead to two publications to international conferences as well as the submission of an additional article

The purpose of this research was to investigate the relationship of small learning communities (SLCs) and student performance for ninth grade students at Robert E. Lee High School in North East Independent School District (NEISD). For this study, student performance includes achievement on reading and math Texas Assessment of Knowledge and Skills (TAKS), attendance rates, and number of dropouts. Research included data for years 2002-2003 through 2005-2006 retrieved from the Academic Excellence Indicator System (AEIS). An extensive review of the literature revealed support for implementation of the SLCs model for high school reform. Recent research indicates that student performance will improve if SLCs are fully implemented and supported using a framework, such as Oxley’s five domains for SLCs. The first two questions of this study addressed reading and math TAKS data by ethnic, economically disadvantaged, and special education subpopulations. The third question in this study addressed attendance rates and dropouts. This study found a significant difference in reading TAKS scores for the ethnic subpopulations with the implementation of SLCs. Economically disadvantaged students’ and special education students’ scores also showed significant gains in reading scale scores over the four years of the study. Data from this study revealed that math TAKS scale scores showed a significant increase in the economically disadvantaged and special education subpopulations after implementation of the SLCs. In addition, significance was found in reducing the achievement gap between special education and regular education students on math TAKS. Attendance rates showed no statistical significance after the implementation of the SLCs. An analysis of dropout rates was not possible due to low dropout numbers. The empirical data would not support meaningful analysis. Further investigation is needed to gain a better understanding of the relationship of SLCs on student performance, especially for African American and Hispanic students in math. Additional factors such as degree of implementation and influence of the administrative leadership needs to be explored.

Le sommeil est connu pour réguler plusieurs fonctions importantes pour le cerveau et parmi celles-ci, il y a le blocage de l’information sensorielle par le thalamus et l’amélioration de la consolidation de la mémoire. Le sommeil à ondes lentes, en particulier, est considéré être critique pour ces deux processus. Cependant, leurs mécanismes physiologiques sont inconnus. Aussi, la marque électrophysiologique distinctive du sommeil à ondes lentes est la présence d’ondes lentes de grande amplitude dans le potentiel de champ cortical et l’alternance entre des périodes d’activités synaptiques intenses pendant lesquelles les neurones corticaux sont dépolarisés et déchargent plusieurs potentiels d’action et des périodes silencieuses pendant lesquelles aucune décharge ne survient, les neurones corticaux sont hyperpolarisés et très peu d’activités synaptiques sont observées. Tout d'abord, afin de mieux comprendre les études présentées dans ce manuscrit, une introduction générale couvrant l'architecture du système thalamocortical et ses fonctions est présentée. Celle-ci comprend une description des états de vigilance, suivie d'une description des rythmes présents dans le système thalamocortical au cours du sommeil à ondes lentes, puis par une description des différents mécanismes de plasticité synaptique, et enfin, deux hypothèses sur la façon dont le sommeil peut affecter la consolidation de la mémoire sont présentées. Puis, trois études sont présentées et ont été conçues pour caractériser les propriétés de l'oscillation lente du sommeil à ondes lentes. Dans la première étude (chapitre II), nous avons montré que les périodes d'activité (et de silence) se produisent de façon presque synchrone dans des neurones qui ont jusqu'à 12 mm de distance. Nous avons montré que l'activité était initiée en un point focal et se propageait rapidement à des sites corticaux voisins. Étonnamment, le déclenchement des états silencieux était encore plus synchronisé que le déclenchement des états actifs. L'hypothèse de travail pour la deuxième étude (chapitre III) était que les états actifs sont générés par une sommation de relâches spontanées de médiateurs. Utilisant différents enregistrements à la fois chez des animaux anesthésiés et chez d’autres non-anesthésiés, nous avons montré qu’aucune décharge neuronale ne se produit dans le néocortex pendant les états silencieux du sommeil à ondes lentes, mais certaines activités synaptiques peuvent ii être observées avant le début des états actifs, ce qui était en accord avec notre hypothèse. Nous avons également montré que les neurones de la couche V étaient les premiers à entrer dans l’état actif pour la majorité des cycles, mais ce serait ainsi uniquement pour des raisons probabilistes; ces cellules étant équipées du plus grand nombre de contacts synaptiques parmi les neurones corticaux. Nous avons également montré que le sommeil à ondes lentes et l’anesthésie à la kétamine-xylazine présentent de nombreuses similitudes. Ayant utilisé une combinaison d'enregistrements chez des animaux anesthésiés à la kétamine-xylazine et chez des animaux non-anesthésiés, et parce que l'anesthésie à la kétamine-xylazine est largement utilisée comme un modèle de sommeil à ondes lentes, nous avons effectué des mesures quantitatives des différences entre les deux groupes d'enregistrements (chapitre IV). Nous avons trouvé que l'oscillation lente était beaucoup plus rythmique sous anesthésie et elle était aussi plus cohérente entre des sites d’enregistrements distants en comparaison aux enregistrements de sommeil naturel. Sous anesthésie, les ondes lentes avaient également une amplitude plus grande et une durée plus longue par rapport au sommeil à ondes lentes. Toutefois, les ondes fuseaux (spindles) et gamma étaient également affectées par l'anesthésie. Dans l'étude suivante (Chapitre V), nous avons investigué le rôle du sommeil à ondes lentes dans la formation de la plasticité à long terme dans le système thalamocortical. À l’aide de stimulations pré-thalamiques de la voie somatosensorielle ascendante (fibres du lemnisque médial) chez des animaux non-anesthésiés, nous avons montré que le potentiel évoqué enregistré dans le cortex somatosensoriel était augmenté dans une période d’éveil suivant un épisode de sommeil à ondes lentes par rapport à l’épisode d’éveil précédent et cette augmentation était de longue durée. Nous avons également montré que le sommeil paradoxal ne jouait pas un rôle important dans cette augmentation d'amplitude des réponses évoquées. À l’aide d'enregistrements in vitro en mode cellule-entière, nous avons caractérisé le mécanisme derrière cette augmentation et ce mécanisme est compatible avec la forme classique de potentiation à long terme, car il nécessitait une activation à la fois les récepteurs NMDA et des récepteurs AMPA, ainsi que la présence de calcium dans le neurone post-synaptique. iii La dernière étude incluse dans cette thèse (chapitre VI) a été conçue pour caractériser un possible mécanisme physiologique de blocage sensoriel thalamique survenant pendant le sommeil. Les ondes fuseaux sont caractérisées par la présence de potentiels d’action calcique à seuil bas et le calcium joue un rôle essentiel dans la transmission synaptique. En utilisant plusieurs techniques expérimentales, nous avons vérifié l'hypothèse que ces potentiels d’action calciques pourraient causer un appauvrissement local de calcium dans l'espace extracellulaire ce qui affecterait la transmission synaptique. Nous avons montré que les canaux calciques responsables des potentiels d’action calciques étaient localisés aux synapses et que, de fait, une diminution locale de la concentration extracellulaire de calcium se produit au cours d’un potentiel d’action calcique à seuil bas spontané ou provoqué, ce qui était suffisant pour nuire à la transmission synaptique. Nous concluons que l'oscillation lente est initiée en un point focal et se propage ensuite aux aires corticales voisines de façon presque synchrone, même pour des cellules séparées par jusqu'à 12 mm de distance. Les états actifs de cette oscillation proviennent d’une sommation de relâches spontanées de neuromédiateurs (indépendantes des potentiels d’action) et cette sommation peut survenir dans tous neurones corticaux. Cependant, l’état actif est généré plus souvent dans les neurones pyramidaux de couche V simplement pour des raisons probabilistes. Les deux types d’expériences (kétamine-xylazine et sommeil à ondes lentes) ont montré plusieurs propriétés similaires, mais aussi quelques différences quantitatives. Nous concluons également que l'oscillation lente joue un rôle essentiel dans l'induction de plasticité à long terme qui contribue très probablement à la consolidation de la mémoire. Les ondes fuseaux, un autre type d’ondes présentes pendant le sommeil à ondes lentes, contribuent au blocage thalamique de l'information sensorielle.
Sleep is known to mediate several major functions in the brain and among them are the gating of sensory information during sleep and the sleep-related improvement in memory consolidation. Slow-wave sleep in particular is thought to be critical for both of these processes. However, their physiological mechanisms are unknown. Also, the electrophysiological hallmark of slow-wave sleep is the presence of large amplitude slow waves in the cortical local field potential and the alternation of periods of intense synaptic activity in which cortical neurons are depolarized and fire action potentials and periods of silence in which no firing occurs, cortical neurons are hyperpolarized, and very little synaptic activities are observed. First, in order to better understand the studies presented in this manuscript, a general introduction covering the thalamocortical system architecture and function is presented, which includes a description of the states of vigilance, followed by a description of the rhythms present in the thalamocortical system during slow-wave sleep, then by a description of the mechanisms of synaptic plasticity, and finally two hypotheses about how sleep might affect the consolidation of memory are presented. Then, three studies are presented and were designed to characterize the properties of the sleep slow oscillation. In the first study (Chapter II), we showed that periods of activity (and silence) occur almost synchronously in neurons that are separated by up to 12 mm. The activity was initiated in a focal point and rapidly propagated to neighboring sites. Surprisingly, the onsets of silent states were even more synchronous than onsets of active states. The working hypothesis for the second study (Chapter III) was that active states are generated by a summation of spontaneous mediator releases. Using different recordings in both anesthetized and non-anesthetized animals, we showed that no neuronal firing occurs in the neocortex during silent states of slow-wave sleep but some synaptic activities might be observed prior to the onset of active states, which was in agreement with our hypothesis. We also showed that layer V neurons were leading the onset of active states in most of the cycles but this would be due to probabilistic reasons; these cells being equipped with the most numerous synaptic contacts among cortical neurons. We also showed that slow-wave sleep and ketamine-xylazine shares many similarities. v Having used a combination of recordings in ketamine-xylazine anesthetized and non-anesthetized animals, and because ketamine-xylazine anesthesia is extensively used as a model of slow-wave sleep, we made quantitative measurements of the differences between the two groups of recordings (Chapter IV). We found that the slow oscillation was much more rhythmic under anesthesia and it was also more coherent between distant sites as compared to recordings during slow-wave sleep. Under anesthesia, slow waves were also of larger amplitude and had a longer duration as compared to slow-wave sleep. However, spindles and gamma were also affected by the anesthesia. In the following study (Chapter V), we investigated the role of slow-wave sleep in the formation of long-term plasticity in the thalamocortical system. Using pre-thalamic stimulations of the ascending somatosensory pathway (medial lemniscus fibers) in non-anesthetized animals, we showed that evoked potential recorded in the somatosensory cortex were enhanced in a wake period following a slow-wave sleep episode as compared to the previous wake episode and this enhancement was long-lasting. We also showed that rapid eye movement sleep did not play a significant role in this enhancement of response amplitude. Using whole-cell recordings in vitro, we characterized the mechanism behind this enhancement and it was compatible with the classical form of long-term potentiation, because it required an activation of both NMDA and AMPA receptors as well as the presence of calcium in the postsynaptic neuron. The last study included in this thesis (Chapter VI) was designed to characterise a possible physiological mechanism of thalamic sensory gating occurring during sleep. Spindles are characterized by the presence of low-threshold calcium spikes and calcium plays a critical role in the synaptic transmission. Using several experimental techniques, we verified the hypothesis that these calcium spikes would cause a local depletion of calcium in the extracellular space which would impair synaptic transmission. We showed that calcium channels responsible for calcium spikes were co-localized with synapses and that indeed, local extracellular calcium depletion occurred during spontaneous or induced low-threshold calcium spike, which was sufficient to impair synaptic transmission. We conclude that slow oscillation originate at a focal point and then propagate to neighboring cortical areas being almost synchronous even in cells located up to 12 mm vi apart. Active states of this oscillation originate from a summation of spike-independent mediator releases that might occur in any cortical neurons, but happens more often in layer V pyramidal neurons simply due to probabilistic reasons. Both experiments in ketamine-xylazine anesthesia and non-anesthetized animals showed several similar properties, but also some quantitative differences. We also conclude that slow oscillation plays a critical role in the induction of long-term plasticity, which very likely contributes to memory consolidation. Spindles, another oscillation present in slow-wave sleep, contribute to the thalamic gating of information.
Tableau d'honneur de la FÉSP

This thesis investigates the Slow Food movement which started in Italy in the late80’s and today is an internationally spread organization with around 100 000 participants in 160 countries. The core aim within the Slow Food movement is combining everyone’s “right to pleasure” with social responsibility, summed up in their motto: “Good, Clean and Fair”. The movement is interesting from a global development perspective, since it looks at food in relation to the cultural, physical, social, environmental and political aspects of our lives, and relates to sustainability in a wide sense. In 2003, UNESCO started acknowledging so called ”Intangible cultural heritage” as a part of the common heritage of humanity, and a few years later some culinary traditions were inserted to the list. Departing from a social constructivist approach in how knowledge is constructed, combined with Pieterse’s ideas of the relation between culture and power, the key research question for the thesis is: How is the concept of Food as cultural heritage being expressed within the external communication channels of the Slow Food movement? Followed by the sub-question: How do these expressions relate to power dynamicsand to socially constructed assumptions of reality? Through a semiotic analysis, visual and textual material from Slow Food’s official websites have been analysed, using the UNESCO definition of Intangible cultural heritage as an analytical tool. The findings are that many fundamental ideas within the UNESCO definition are expressed in Slow Food’s external communication, but at the same time, that some aspects could be more highlighted in order to put a higher emphasis on the producer and the community. It is evident that Slow Food need to continuously address the critique directed towards them regarding issues of privilege and elitism, and keep a self-reflexive approach in their communication work.

P2P live streaming applications are becoming more popular each day. In contrast to the client-server model, the P2P model overcomes problems like system scalability and need of powerful resources in a single point. On the other hand, P2P networks may suffer with attacks and opportunistic behavior. In this work we present a decentralized reputation system to fight attacks in P2P live streaming networks. In particular, we focus on pollution attacks, where a malicious peer alters or forges the streaming data, aiming to disseminate undesirable content. Our results include a formal model to check pollution attacks damages to P2P live streaming systems. We show that protocol and algorithms influence the impact caused by an attack. We also characterize the client behavior in P2P live streaming systems. Thus, we can understand client dynamics and evaluate the impact that it causes on system scalability and media quality. Finally we implement a reputation mechanism to fight pollution attacks. Our results show that the new mechanism can block a pollution attack with an overhead lower than 2% of network bandwidth, while the solution recommend needs at least 100%. Moreover, the new mechanism can lead with collusion attacks. In this case, it needs a peak less than 30%, while traditional approach needs at least 100% more bandwidth. Finally, allows peers rehabilitation and provides incentive mechanisms.
Aplicações de mídia contínua em P2P estão se tornando mais populares a cada dia. Em contraste com o modelo tradicional cliente-servidor, o modelo P2P transpassa problemas como escalabilidade e necessidade de recursos poderosos, concentrados em um único ponto. Em contrapartida, as redes P2P podem ser alvos de ataques e comportamentos oportunistas. Este trabalho apresenta um modelo descentralizado de reputação para combater ataques em sistemas de mídia contínua P2P. Em particular, são concentrados esforços para combater ataques de poluição. Nesse ataque, os participantes maliciosos alteram ou forjam os dados da mídia contínua na tentativa de disseminar conteúdo indesejado aos demais participantes do sistema. Os resultados incluem uma modelagem de sistemas P2P e uma verificação formal dos danos causados por ataques de poluição. Nesse sentido, há uma evidencia que a escolha dos protocolos e algoritmos do sistema P2P influenciam no impacto causado por um ataque. Incluem também uma caracterização dos usuários de sistemas de transmissão ao vivo de mídia contínua em P2P. A partir dessa caracterização é possível compreender a dinâmica dos participantes desses sistemas e assim avaliar o impacto de seu comportamento tanto na escalabilidade do sistema, quanto na qualidade da mídia recebida. Finalmente, é desenvolvido um sistema de defesa contra ataques de poluição. Os resultados mostram que o novo mecanismo implementado bloqueia um ataque com uma sobrecarga inferior a 2% da banda de rede necessária para a transmissão, enquanto a solução indicada pela literatura necessita de pelo menos 100% de banda adicional. Além disso, o novo mecanismo é eficaz mesmo com conluio dos participantes e, nesse caso, o pico de retransmissão é inferior a 30%, enquanto as soluções tradicionais necessitam de até 100% de banda adicional. Finalmente, o novo mecanismo permite reabilitação dos participantes que geraram conteúdo indevido e provê medidas de incentivo para o ompartilhamento dos recursos na rede P2P

This work proposes a security protocol against wormhole attacks in wireless sensor networks (WSN), based in message monitoring by base station and use of alternative paths. WSN are ad hoc networks,comprised of centuries of small sensor devices, used to monitor areas of interest. Its application goes since use in battle fields until environmental protection. The devices, also called sensor nodes, have limited resources for processing and communicating. Data collected by nodes are sent towards the base station, through routes created to this finality. WSN can suffer several types of attacks performed by maliciousnodes or invaders. The wormhole attack generally is accomplished by two colluded nodes. Their strategy is to attract to themselves the most of routs and messages. After that they can act on messages they receivein several ways. The most common action is to drop them, creating a shadow zone in the network, from where no data can be obtained. Solutions based on cryptography or massage authentication are noteffective to avoid it, because the opponent doesnt need to break messages contents to setup the attack. Also this kind of enemy can be superior equipment in terms of computational resources e radio capability compared to the common nodes making its combat a hard task. The solution we propose allows a WSN accomplish its work even during a wormhole attack, with a minimum message loss creating a tolerant attack security protocol. On experiments performed the results were around 90% of gain or higher to 500 nodes networks and attack rate of 70% and over.
Este trabalho propõe um protocolo de segurança para tolerância a ataques de tunelamento em Redes de sensores sem fio (RSSF), baseado no uso de caminhos alternativos e no monitoramento de mensagens realizado pela estação base. As RSSFs são redes ad hoc, compostas por centenas de dispositivos sensores, usadas para monitorar áreas de interesse. Sua aplicação varia desde reconhecimento de campos de batalha até proteção ambiental. Os dispositivos, também chamados de nós sensores, possuem recursos limitados de processamento e comunicação. Os dados coletados pelos nós são enviados à estação base através de rotas criadas para esse fim. As RSSFs podem sofrer diversos tipos de ataques provocados por nós maliciosos ou invasores. O ataque por tunelamento, geralmente, é executado por dois nós da rede, em conluio, cuja estratégia é atrair para si o maior número possível de rotas e mensagens. Posteriormente, os atacantes podem agir sobre essas mensagens de várias formas, sendo a mais comum o seu descarte, criando uma região na rede da qual não se obterá nenhum dado sensoreado. As soluções que utilizam criptografia e autenticação nas mensagens não são suficientes para detê-los, pois estes não necessitam decodificar as mensagens para estabelecer o ataque. Além disso, este tipo de atacante pode ser um equipamento de recursos computacionais e de comunicação muito superiores aos disponíveis para os nós comuns, o que dificulta o seu combate. A solução que apresentamos visa permitir que uma RSSF, mesmo sob um ataque de tunelamento, consiga realizar a sua missão com um mínimo de perda de mensagens, criando-se um certo grau de tolerância ao ataque. Nos experimentos realizados obtivemos resultados com ganho em torno de 90%, ou mesmo superior, para redes de até 500 nós e freqüência de ataques acima de 70% do tempo.

This work addresses the problem of physical route selection for Label Switched Paths (LSPs) in Multi-Protocol Label Switching (MPLS) networks. It consists of defining routes for LSPs trying to minimize the total number of hops and the total number of rejections. The LSPs path selection problem is addressed through two different approaches: on-line and offline. We propose an ILP (Integer Linear Programming) model to solve the offline problem in an exact manner. This model tries to balance the network load while minimizing the network rejection rate. The goal is to optimize the overall network performance by routing requests through under-utilized links. Issues concerning the execution time necessary to solve the model are noticed and a genetic algorithm (GA) is developed to address theses issues. The GA is based on the combination of routing policies and adaptive route movements. In order to analyze the genetic algorithm's performance, various simulations are conducted. Results show that the GA executes much faster than the commercial tool used to solve the model in an exact manner. Results also show that the use of the GA considerably decreases the amount of rejected requests in the network. Experiments are conducted and satisfactory results are found showing that the GA is also suitable for on-line routing. This work also addresses the problem of adjusting the network routing in other cases besides the arrival of new requests. The process of adjusting the network routing is called Grooming and it occurs when a new network element is configured, LSPs are torn down or a network element recovers from failure. It is necessary when routing decisions taken in the past are no longer efficient. A traffic/resource change detection mechanism is proposed to trigger the Grooming process. This mechanism is based on control charts and is used in the context of defining the moment the Grooming should take place. Simulation results indicate that better performance is achieved using the proposed mechanisms.
Esse trabalho trata do problema de seleção de rotas físicas para LSPs (Label Switching Paths) em redes MPLS (Multi-Protocol Label Switching). Esse problema consiste na definição de rotas para os LSPs que minimizem o número total de saltos e o número total de rejeições. O problema é tratado através de duas estratégias diferentes: on-line e offline. Inicialmente propomos um modelo linear inteiro para solucionar o problema de maneira exata. Esse modelo tenta balancear a carga na rede enquanto minimiza a taxa de rejeição. O objetivo é otimizar o desempenho global da rede roteando as requisições através de enlaces sub-utilizados melhorando a utilização da infra-estrutura instalada. Há vários problemas relacionados com o tempo necessário para resolver o modelo e por isso desenvolvemos um algoritmo genético, baseado na combinação de políticas de roteamento e movimentos adaptativos de rotas. Para analisar o desempenho do algoritmo genético, foram estudados vários cenários e foram feitas várias simulações. Os resultados mostram que o algoritmo genético executa muito mais rápido que a ferramenta comercial utilizada para resolver o model de forma exata. Os resultados também mostram que o uso do algoritmo genético diminui consideravelmente a quantidade de requisições rejeitadas na rede. Foram executados vários experimentos e resultados satistatórios foram encontrados, mostrando que o algoritmo genético é também apropriado para roteamento on-line. Além do caso clássico de chegada de novas requisições, esse trabalho também trata do problema de ajuste do roteamento em diversos outros casos. Chamamos de Grooming esse processo de ajuste do roteamento da rede e ele ocorre quando um novo elemento de rede é configurado, LSPs são terminados ou elementos de rede se recuperam de falhas. O Grooming da rede é necessário quando decições de roteamento tomadas no passado não são mais eficientes. Um mecanismo de detecção de mudanças é proposto para engatilhar o processo de Grooming. Esse mecanismo é baseado em cartas de controle e é usado para definir o momento em que o Grooming deve acontecer. Resultados de simulação indicam que um melhor desempenho é alcançado através do uso dos mecanismos propostos.

In this work we propose, implement and evaluate strategies and techniques for the problem of extracting semistructured data fromWeb data sources within the context of an approach we call DEByE (Data Extraction By Example). The results we have reached have been used in the implementation of a data extraction tool, also called DEByE, and have their effectiveness verified through experiments. The DEByE approach is semi-automatic, in the sense that the role of users (i.e., wrapper developers) is limited to providing examples of the data to be extracted, what shields them from being aware of specific formatting features of the target pages. The examples provided describe the structure of the objects being extracted by means of nested tables, which are simple and intuitive, and expressive enough to represent the structure of the data normally present in Web pages. To deal with typical variations of complex semistructured objects, we have extended the original concept of nested tables by relaxing the original assumption that all inner tables nested in a column should have a same internal structure. Based on this extended form of nested tables, we formalize the concept of wrappers by means of tabular grammars. Such context-free grammars are formed by productions that lead to parse trees that can be directly mapped to nested tables. We have developed strategies for generating tabular grammars from a set of example objects provided by a user from a sample page. This includes: (1) the generation of terminal productions for extracting single values belonging to a specific domain (e.g., an item description, a price, etc.) and (2) the generation of non-terminal productions that represent the structure of the complex objects to be extracted. The extraction of data from target pages is accomplished by parsing these pages using a tabular grammar. For this, we have developed an eficient bottom-up strategy. This strategy includes two distinct phases: an extraction phase, in which atomic attribute values are extracted based on local context information available in the extraction productions, and an assembling phase, in which such values are assembled to form complex objects according to the target structure supplied by the user through examples, which is encoded in the non-terminal productions. We experimentally demonstrate the effectiveness of the bottom-up strategy for dealing with multi-level objects presenting structural variations. The general principle used by the bottom-up algorithm, that is, first extracting atomic values and then grouping these values to assemble complex objects, has been further exploited by the Hot Cycles algorithm we have developed. This algorithm aims at uncovering a plausible tabular structure for assembling complex objects with a given set of atomic values extracted from a target page. This algorithm is useful for deploying the DEByE approach in applications where the user is not available for assembling example tables.
Neste trabalho, são propostas, implementadas e avaliadas estratégias e técnicas para o problema de extração de dados semi-estruturados de fontes de dados da Web, dentro do contexto de uma abordagem chamada DEByE (Data Extraction By Example). Os resultados obtidos com o trabalho foram usados na implementação de um ferramenta de extração de dados, também chamada DEByE, e tiveram sua eficácia verificada através de experimentação. A abordagem DEByE é dita semi-automática, no sentido em que o papel dos usuários (ou seja, dos desenvolvedores de extratores) é limitado ao fornecimento de exemplos dos dados a serem extraídos, o que os isola de ter que conhecer as características especifícas de formação das páginas alvo. Os exemplos fornecidos descrevem a estrutura dos objetos a serem extraídos por meio de tabelas aninhadas, as quais são simples, intuitivas e expressivas o suficiente para representar a estrutura dos dados normalmente presentes empáginas Web. Para lidar com variações típicas de objetos complexos semi-estruturados, estendemos o conceito original de tabelas aninhadas, relaxando a restrição original de que as tabelas aninhadas em uma mesma coluna devem todas ter a mesma estrutura interna. Com base nesta forma estendida de tabelas aninhadas, formalizamos o conceito de wrapper através de gramáticas tabulares. Tais gramáticas livres de contexto são compostas por produções que levam a árvores de parsing que podem ser diretamente mapeadas para tabelas aninhadas. Foram desenvolvidas estratégias para gerar gramáticas tabulares a partir de um conjunto de objetos de exemplo fornecidos por um usuário que os seleciona de uma página de amostra. O processo de geração inclui: (1) geração de produções terminais para extrair valores atômicos pertencentes a um domínio específico (por exemplo, uma descrição de um item, um preço, etc.) e (2) a geração de produções não-terminais que representam a estrutura de objetos complexos a serem extraídos. A extração dos dados das páginas alvo é feita através de um processo de parsing sobre estas páginas usando a gramática tabular. Para isso, desenvolvemos uma eficiente estratégia bottom-up, a qual envolve duas fases distintas: uma fase de extração, na qual os valores atômicos dos atributos são extraídos como base na informação de contexto local disponível nas produções de extração, e uma fase de montagem, na qual os valores atômicos extraídos são montados de acordo com a estrutura alvo fornecida pelo usuário através de exemplos e que está representada nas produções não terminais. A eficácia da estratégia bottom-up foi comprovada experimentalmente, em especial para tratar objetos complexosmulti-nível que apresentam variações estruturais. O principio geral utilizado pela estratégia bottom-up, ou seja, primeiro extrair valores atômicos e depois agrupar estes valores para montar objetos complexos, foi também explorado por um outro algoritmo que desenvolvemos chamado Hot Cycles. Este algoritmo tem como objetivo descobrir estruturas tabulares plausíveis para montar objetos complexos com um conjunto de valores atômicos extraídos de uma página alvo. Ele é útil para o emprego da abordagem DEByE em aplicações onde não se pode depender do usuário para montar tabelas de exemplo.