In many relevant scenarios, there is an imbalance between abundant unlabeled data and scarce labeled data to train predictive models. Semi-Supervised Learning and Active Learning are two distinct approaches to deal with this issue. The first one directly uses the unlabeled data to improve model parameter learning, while the second performs a smart choice of unlabeled points to be sent to an annotator, or oracle, which can label these points and increase the labeled training set. In this context, Generative Models are highly appropriate, since they internally represent the data generating process, naturally benefiting from data samples independently of the presence of labels. This Thesis proposes Expectation-Maximization with Density-Weighted Entropy, a novel active semi-supervised learning framework tailored towards generative models. The method is theoretically explored and experiments are conducted to evaluate its application to Gaussian Mixture Models and Multinomial Mixture Models. Based on its partial success, several questions are raised and discussed as to identify possible improvements and decide which shortcomings need to be dealt with before the method is considered robust and generally applicable. / I många relevanta scenarier finns det en obalans mellan god tillgång på oannoterad data och sämre tillgång på annoterad data för att träna prediktiva modeller. Semi-Övervakad Inlärning och Aktiv Inlärning är två distinkta metoder för att hantera denna fråga. Den första använder direkt oannoterad data för att förbättra inlärningen av modellparametrar, medan den andra utför ett smart val av oannoterade punkter som ska skickas till en annoterare eller ett orakel, som kan annotera dessa punkter och öka det annoterade träningssetet. I detta sammanhang är Generativa Modeller mycket lämpliga eftersom de internt representerar data-genereringsprocessen och naturligt gynnas av dataexempel oberoende av närvaron av etiketter. Denna Masteruppsats föreslår Expectation-Maximization med Density-Weighted Entropy, en ny aktiv semi-övervakad inlärningsmetod som är skräddarsydd för generativa modeller. Metoden utforskas teoretiskt och experiment genomförs för att utvärdera dess tillämpning på Gaussiska Mixturmodeller och Multinomiala Mixturmodeller. Baserat på dess partiella framgång ställs och diskuteras flera frågor för att identifiera möjliga förbättringar och avgöra vilka brister som måste hanteras innan metoden anses robust och allmänt tillämplig.

Implicit communication is important in interaction because it plays a role in conveying the internal mental states of an individual. For example, emotional expressions that are shown through unintended facial gestures can communicate underlying affective states. People can infer mental states from implicit cues and have strong expectations of what those cues mean. This is true for human-human interactions, as well as human-robot interactions. A Normalizing flow model is used as a generative model that can produce facial gestures and head movements. The invertible nature of the Normalizing flow model makes it possible to manipulate attributes of the generated gestures. The model in this work is capable of generating facial expressions that look real and human-like. Furthermore, the model can manipulate the generated output to change the perceived affective state of the facial expressions. / Implicit kommunikation är viktig i interaktioner eftersom den spelar en roll för att förmedla individens inre mentala tillstånd. Till exempel kan känslomässiga uttryck som visas genom oavsiktliga ansiktsgester kommunicera underliggande affektiva tillstånd. Människor kan härleda mentala tillstånd från implicita ledtrådar och har starka förväntningar på vad dessa ledtrådar betyder. Detta gäller för interaktion mellan människor, liksom interaktion mellan människa och robot. En normaliserande flödesmodell används som en generativ modell som kan producera ansiktsgester och huvudrörelser. Den inverterbara naturen hos normaliseringsflödesmodellen gör det också möjligt att manipulera det genererade ansiktsuttrycken. Utgången manipuleras i två dimensioner som vanligtvis används för att beskriva affektivt tillstånd, valens och upphetsning. Modellen i detta arbete kan generera ansiktsuttryck som ser verkliga och mänskliga ut och kan manipuleras for att ändra det affektiva tillstånd.

The field of Deep Learning (DL) only over the last decade has proven useful for increasingly more complex Machine Learning tasks and data, a notable milestone being generative models achieving facial synthesis indistinguishable from real faces. With the increased complexity in DL architecture and training data, follows a steep increase in time and hardware resources required for the training task. These resources are easily accessible via cloud-based platforms if the data owner is willing to share its training data. To allow for cloud-sharing of its training data, The Swedish Transport Administration (TRV) is interested in evaluating resource effective, infrastructure independent, privacy-preserving obfuscation methods to be used on real-time collected data on distributed Internet-of-Things (IoT) devices. A fundamental problem in this setting is to balance the trade-off between privacy and DL utility of the obfuscated training data. We identify statistically measurable relevant metrics of privacy achievable via obfuscation and compare two prominent alternatives from the literature, optimization-based methods (OBM) and random projections (RP). OBM achieve privacy via direct optimization towards a metric, preserving utility-crucial patterns in the data, and is typically in addition evaluated in terms of a DL-based adversary’s sensitive feature estimation error. RP project data via a random matrix to lower dimensions to preserve sample pair-wise distances while offering privacy in terms of difficulty in data recovery. The goals of the project centered around evaluating RP on privacy metric results previously attained for OBM, compare adversarial feature estimation error in OBM and RP, as well as to address the possibly infeasible learning task of using composite multi-device datasets generated using independent projection matrices. The last goal is relevant to TRV in that multiple devices are likely to contribute to the same composite dataset. Our results complement previous research in that they indicate that both privacy and utility guarantees in a distributed setting, vary depending on data type and learning task. These results favor OBM that theoretically should offer more robust guarantees. Our results and conclusions would encourage further experimentation with RP in a distributed setting to better understand the influence of data type and learning task on privacy-utility, target-distributed data sources being a promising starting point. / Forskningsområdet Deep Learning (DL) bara under det senaste decenniet har visat sig vara användbart för allt mer komplexa maskinginlärnings-uppgifter och data, en anmärkningsvärd milstolpe är generativa modeller som erhåller verklighetstrogna syntetiska ansiktsbilder. Med den ökade komplexiteten i DL -arkitektur och träningsdata följer ett kraftigt ökat behov av tid och hårdvaruresurser för träningsuppgiften. Dessa resurser är lättillgängliga via molnbaserade plattformar om dataägaren är villig att dela sin träningsdata. För att möjliggöra molndelning av träningsdata är Trafikverket (TRV) intresserat av att utvärdera resurseffektiva, infrastrukturoberoende, privatiserade obfuskeringsmetoder som ska användas på data hämtad i realtid via distribuerade Internet-of-Things ( IoT) -enheter; det grundläggande problemet är avvägningen mellan privatisering och användbarhet av datan i DL-syfte. Vi identifierar statistiskt mätbara relevanta mått av privatisering som kan uppnås via obfuskering och jämför två framstående alternativ från litteraturen, optimeringsbaserade metoder (OBM) och slumpmässiga projektioner (RP). OBM uppnår privatisering via matematisk optimering av ett mått av data-privatisering, vilket bevarar övriga nödvändiga mönster i data för DL-uppgiften. OBM-metoder utvärderas vanligtvis i termer av en DL-baserad motståndares uppskattningsfel av känsliga attribut i datan. RP obfuskerar data via en slumpmässig projektion till lägre dimensioner för att bevara avstånd mellan datapunkter samtidigt som de erbjuder privatisering genom teoretisk svårighet i dataåterställning. Målen för examensarbetet centrerades kring utvärdering av RP på privatiserings-mått som tidigare uppnåtts för OBM, att jämföra DL-baserade motståndares uppskattningsfel på data från OBM och RP, samt att ta itu med den befarat omöjliga inlärningsuppgiften att använda sammansatta dataset från flera IoT-enheter som använder oberoende projektionsmatriser. Sistnämnda målet är relevant i en miljö sådan som TRVs, där flera IoT-enheter oberoende bidrar till ett och samma dataset och DL-uppgift. Våra resultat kompletterar tidigare forskning genom att de indikerar att både privatisering och användbarhetsgarantier i en distribuerad miljö varierar beroende på datatyp och inlärningsuppgift. Dessa resultat gynnar OBM som teoretiskt sett bör erbjuda mer robusta garantier vad gäller användbarhet. Våra resultat och slutsatser uppmuntrar framtida experiment med RP i en distribuerad miljö för att bättre förstå inverkan av datatyp och inlärningsuppgift på graden av privatisering, datakällor distribuerade baserat på klassificerings-target är en lovande utgångspunkt.

AI-generativa bildkonst verktyg använder sig av generative adversarial networks (GAN) för att skapa bilder. Generative adversarial networks blir desto bättre ju mer information den förses med. Denna studie undersöker hur teknologiskt accepterat det är med AI-genererade bildkonst verktyg med hjälp av modellen Technology Acceptance Model (TAM). Syftet med studien är att undersöka om unga konstnärers syn och inställning på AI-genererad bildkonst och verktygets framfart samt om de anser att det är accepterat. Studien genomfördes med små n-studier med semistrukturerade intervjuer för att få en rik och detaljerad beskrivning om acceptansen kring AI-generativa bildkonst verktyg. Studiens intervjufrågor byggdes upp för att besvara TAM2s olika delar. Respondenterna var konstnärer från Sverige och valdes ut genom bekvämlighetsurval kombinerat med snowball sampling. Konstnärerna som intervjuades under studien var positivt inställda till att använda AI-bildverktygen som inspiration och hjälpmedel till deras egna konst, och de flesta konstnärerna ansåg inte att AI var ett hot för dem eftersom de höll på med fysiska konstverk, och såg AI mer som ett hot för digitala konstnärer. Dessutom tyckte de flesta konstnärerna i studien att även om det är en själv som har skrivit in instruktionerna till AI-bildverktygen för att få en genererad bild, så är det diskutabelt om man kan göra anspråk på verket. / AI generated visual art tools use generative adversarial networks (GAN) to create pictures and digital art. Generative adversarial networks get better the more information they are provided with. This study investigates how technologically accepted it is with AI generated visual art tools using the Technology Acceptance Model (TAM). This study was carried out with small n-studies with semi-structured interviews to get rich and detailed description of the acceptance of AI generative visual art tools. The study’s interview questions were built to answer the TAM2-model different parts. The respondents were artists from Sweden and were selected through convenience sampling combined with snowball sampling. The purpose of the study is to investigate how young artists view and attitude towards AI generated visual art and the progress of the tool and whether they believe it is accepted. The artists interviewed during the study were positive about using the AI generated art tool as an inspiration and aid to their own art, and most of the artists did not consider AI as a threat to them because they were doing physical artwork, seeing AI more as a threat to digital artists. In addition, most of the artists in the study felt that even if one has entered the instructions to the AI generating tools themselves to get a generated image, it is debatable whether one can claim the work. This thesis is written in Swedish.

The integration of artificial intelligence (AI) into biomedical research has given rise to new models and research topics in biomedicine. Whole-cell modeling aims to create a holistic understanding of the cell by integrating diverse data. One method of comprehension is the characterization and imitation of a system. Phenomenological cell models imitate cell structure and behavior based on, for example, images. Thus generative AI image models present one approach to developing such phenomenological models of cell systems. Diffusion models are a popular generative model class for image generation. Briefly, diffusion models consist of a forward and reverse diffusion process, where the forward process iteratively adds noise to an image and the reverse process learns to remove it. Image generation is achieved by sampling from noise and applying the learned reverse process. The generation may be conditioned to achieve a specific output. The diffusion process is computationally expensive to evaluate in pixel space. The latent diffusion model presents a solution by moving the diffusion process to the latent space of an autoencoder. A latent diffusion model has been trained to develop a phenomenological model of cells in mitosis. The aim is to identify spatial and temporal patterns in the dataset, consisting of fluorescence microscopy images of cells in mitosis, and condition the output of the latent diffusion model on labels associated with the data. The latent diffusion can generate images unconditionally and conditionally. The unconditionally generated images appear visually similar, but quantitative metrics suggest the potential for improvement. Qualitative analysis of the conditionally generated images indicates opportunities for enhancement. The analysis from the proposed method for objective assessment of conditionally generated images, feature extraction of images followed by dimension reduction using uniform manifold approximation and projection, concurs with the visual assessment. However, the quantitative metrics and the proposed method of conditional assessment rely upon InceptionV3 to extract features from the images. InceptionV3 has not been trained on biomedical images and thus the metrics and methods should not be overly relied upon. In general, there is a need for new assessment techniques suitable for non-class conditionally generated images that are unsuitable for evaluation using user studies. / Integrering av artificiell intelligens (AI) i biomedicinsk forskning har gett upphov till nya modeller och forskningsfrågor inom biomedicin. Helcellsmodellering syftar till att skapa ett kvantitativt perspektiv på cellbiologi och skapa holistisk kunskap om cellen. Ett system kan förstås genom karaktärisering och imitation. Generativ AI är ett tillvägagångssätt för att utveckla modeller som kan imitera och karaktärisera celler baserat på bilder. Diffusionsmodeller är en populär klass av generativa modeller för bildgenerering. Diffusionsmodeller består av en framåt- och bakåtdiffusionsprocess, där den framåtriktade processen iterativt lägger till brus i en bild och den bakåtriktade processen lär sig att ta bort det. Nya bilder genereras genom att tillämpa den inlärda bakåtriktade processen på en bild av brus. Generationen kan göras villkorlig för att forma bilden efter givna villkor. Den beräkningsintensiva diffusionsprocessen kan effektiviseras genom att introducera en "autoencoder" som flyttar diffusionsprocessen från pixelrummets stora dimension till det latenta rummet, som har en mindre dimension. Det utgör basen för en latent diffusionsmodell. För att utveckla en fenomenologisk modell av celler i mitos har en latent diffusionsmodell tränats på fluorescensmikroskopibilder på celler som genomgår mitos. Målet är att identifiera spatiala och temporala mönster i bilderna och skapa en modell som kan villkora bildgenerationen baserat på givna spatiala och temporala villkor associerade med bilderna. Latenta diffusionsmodeller kan skapa bilder både villkorligen och helt fritt från den underliggande datadistributionen. Den fria generationen av bilder resulterar i visuellt lika bilder men kvantitativa mått indikerar att modellen kan förbättras. Villkorligt genererade bilder håller inte samma visuella kvalité. Behovet av tekniker för att utvärdera villkorligt genererade bilder har identifierats och en metod har föreslagits. Metoden involverar att extrahera attribut från bilderna och reducera dimensionen av attributen för att visualisera de olika villkoren. Utvärderingen av de villkorligt genererade bilderna visar att den villkorliga generationen kan förbättras. Däremot beror metoden och de kvantitativa mått som beräknades för den fria generationen av bilder på ett neuralt nätverk som inte tränats på biomedicinska bilder. Därför bör resultaten tolkas med viss reservation.

4D Imaging Radar is increasingly becoming a critical component in various industries due to beamforming technology and hardware advancements. However, it does not replace visual data in the form of 2D images captured by an RGB camera. Instead, 4D radar point clouds are a complementary data source that captures spatial information and velocity in a Doppler dimension that cannot be easily captured by a camera's view alone. Some discriminative features of the scene captured by the two sensors are hypothesized to have a shared representation. Therefore, a more interpretable visualization of the radar output can be obtained by learning a mapping from the empirical distribution of the radar to the distribution of images captured by the camera. To this end, the application of deep generative models to generate images conditioned on 4D radar data is explored. Two approaches that have become state-of-the-art in recent years are tested, generative adversarial networks and diffusion models. They are compared qualitatively through visual inspection and by two quantitative metrics: mean squared error and object detection count. It is found that it is easier to control the generative adversarial network's generative process through conditioning than in a diffusion process. In contrast, the diffusion model produces samples of higher quality and is more stable to train. Furthermore, their combination results in a hybrid sampling method, achieving the best results while simultaneously speeding up the diffusion process. / 4D bildradar får en alltmer betydande roll i olika industrier tack vare utveckling inom strålformningsteknik och hårdvara. Det ersätter dock inte visuell data i form av 2D-bilder som fångats av en RGB-kamera. Istället utgör 4D radar-punktmoln en kompletterande datakälla som representerar spatial information och hastighet i form av en Doppler-dimension. Det antas att vissa beskrivande egenskaper i den observerade miljön har en abstrakt representation som de två sensorerna delar. Därmed kan radar-datan visualiseras mer intuitivt genom att lära en transformation från fördelningen över radar-datan till fördelningen över bilderna. I detta syfte utforskas tillämpningen av djupa generativa modeller för bilder som är betingade av 4D radar-data. Två metoder som har blivit state-of-the-art de senaste åren testas: generativa antagonistiska nätverk och diffusionsmodeller. De jämförs kvalitativt genom visuell inspektion och med kvantitativa metriker: medelkvadratfelet och antalet korrekt detekterade objekt i den genererade bilden. Det konstateras att det är lättare att styra den generativa processen i generativa antagonistiska nätverk genom betingning än i en diffusionsprocess. Å andra sidan är diffusionsmodellen stabil att träna och producerar generellt bilder av högre kvalité. De bästa resultaten erhålls genom en hybrid: båda metoderna kombineras för att dra nytta av deras respektive styrkor. de identifierade begränsningarna i de enskilda modellerna och kurera datan för att jämföra hur dessa modeller skalar med större datamängder och mer variation.

A reutilização de software busca aumentar a qualidade e produtividade no desenvolvimento de software, evitando a duplicação do esforço e reaproveitando o máximo possível das experiências de projetos passados. Apesar de simples, esta idéia não é facilmente colocada em prática, principalmente de maneira sistemática e controlada. Técnicas de engenharia de domínio e linhas de produtos de software buscam facilitar esta tarefa, porém ainda existem outros fatores que dificultam a adoção da prática da reutilização. Entre estes, destacam-se os problemas inerentes ao desenvolvimento de software da maneira como é conduzido atualmente, baseado em código-fonte. Estes problemas têm suas origens na crescente demanda por software cada vez mais complexo e afetam negativamente a capacidade de reutilizar software. O desenvolvimento orientado a modelos surge como uma alternativa atraente neste cenário, elevando a importância de modelos dentro do ciclo de vida do software, incorporando-os como parte integrante do produto final por meio de técnicas de modelagem e geração de código. Com isto, parte da complexidade do software fica escondida dentro dos geradores, protegendo os desenvolvedores, reduzindo a incidência de erros, aumentando a produtividade, qualidade, interoperabilidade e manutenibilidade dos artefatos produzidos. Nesta dissertação defende-se a tese de que o desenvolvimento orientado a modelos pode efetivamente aumentar e/ou melhorar a reutilização de software, e que para isso ela deve ser tratada de forma consistente dentro de um processo de engenharia de domínio. Para demonstrar esta tese, é apresentada uma abordagem orientada a modelos para reutilização de software, com atividades que guiam o desenvolvedor durante a análise, projeto e implementação do domínio. São também apresentados os resultados de uma avaliação envolvendo três estudos empíricos, realizados em ambiente acadêmico e industrial, que buscou determinar a viabilidade da abordagem e os benefícios que podem ser alcançados com a combinação de técnicas do desenvolvimento orientado a modelos e da reutilização de software. Os resultados mostram que a abordagem pode trazer diferentes benefícios para organizações de software, incluindo aumento da quantidade e qualidade da reutilização, e reduzindo a complexidade de desenvolvimento e configuração de produtos / Software reuse aims at increasing quality and productivity in software development, avoiding effort duplication and reusing all past experiences possible. Although it is a simple idea, it is not easy to put reuse in practice, especially in a systematic and controlled way. Domain engineering and software product lines techniques try to make this task easier, but there are many other factors that difficult the reuse adoption. Among these factors are the problems that are inherent to software development in the way it is conducted today, based on source code. These problems arise from the growing demand for increasingly complex software, negatively affecting the ability to reuse. Model-driven development is an attractive alternative in this scenario, leveraging the importance of models in the software life cycle, incorporating them as part of the final product through modeling and code generation techniques. As a result, part of the software complexity becomes hidden inside the generators, shielding the developers, reducing errors, increasing the productivity, quality, interoperability and maintainability of the produced assets. In this dissertation is presented the thesis that model-driven development can effectively increase and/or improve software reuse, and that to achieve this goal it must be treated in a consistent way inside a domain engineering process. To demonstrate this thesis, a model-driven software reuse approach is presented, with activities that guide the developer during domain analysis, design and implementation. The results of an evaluation involving three empirical studies are also presented. The studies were performed in both academic and industrial environments, and aimed at determining the viability of the approach and the benefits that can be achieved with the combination of model-driven development and software reuse techniques. The results showed that the approach can bring different benefits to software organizations, such as software reuse quantity and quality improvements, and complexity reduction in product development and configuration tasks

Made available in DSpace on 2016-04-29T14:23:48Z (GMT). No. of bitstreams: 1 Luis Claudio Costa Fajardo.pdf: 2133826 bytes, checksum: 2ce40e62f84603c57f9022d492da0d44 (MD5) Previous issue date: 2009-04-23 / The project of dissertation Le Livre and the Sintext : The simulation of Mallarmé s dream through the digital poetics of Pedro Barbosa intends to evidence the connection between these two works, apart in time line but very intimate considering their poetic ideals: the unfinished Mallarmé s work Le Livre , ideated on XIX century and the digital text synthesizer, Sintext , anthological cyberliterary project, conceived by Pedro Barbosa and Manuel Torres in 2001. Mallarmé s masterpiece, the poem Dice Thrown Never Will Annul Chance , is probably the closest poetic experience of the Livre . This poem is, first of all, a poetic process, which the verses, distant one from the other and printed in various typefaces lead to a not lineal and limitless reading that offers the reader many entries and exits. The research approaches, at first, the development of cyber literature from hypertext s paradigm by going trough the multimedia digital poetics, the ergodic literature until focusing literary occurrences generated from a computer: the digital poetry, the generative literature and the hyperfiction, according to Pedro Barbosa´s proposal. In this trajectory, authors like Jay David Bolter, George Landow, Jacques Derrida. Levi Manovich, Lúcia Santaella, Giselle Bielguelman, Janet Murray, Aspen Arseth,Pedro Barbosa, Chris Funkenhouser, Rui Torres e Jorge Luiz Antônio are related. The dissertation proceeds with a presentation of the French poet s poetic work, and it proposes a comparison between the Sintext and the Livre , by coming into view the coincident points between these works, with the support of the work of the authors Maurice Blanchot , Jacques Scherer, Arlindo Machado, Octavio Paz, beside the others before mentioned. The climax of the project is a poetic experimentation in a digital medium wich Mallarmé s poem Dice Thrown Never Will Annul Chance is submitted to a text engine similar to the Sintext . The process is put into practice by a combinatorial operation with the verses of poem; it will enable innumerable possibilities of re-combining the verses, amplifying the meaning of the work and playing, by the digital poetics, the multiple and infinite book once conceived by Mallarmé / O projeto de dissertação, Le Livre e o Sintext : A simulação do sonho de Mallarmé através da Poética Digital de Pedro Barbosa, pretende evidenciar a relação existente entre duas obras distantes no tempo, porém íntimas em seus ideais poéticos: a obra inacabada de Mallarmé conhecida como Le Livre , idealizada no século XIX e o sintetizador de textos em meio digital, o Sintext , projeto antológico da ciberliteratura, concebida por Pedro Barbosa e José Manuel Torres em 2001. Acredita-se que o poema, Um lance de dados , obra-prima de Mallarmé, tenha sido a experiência poética mais próxima do Livre . Tal poema é, antes de tudo, um processo poético, em que os versos, distantes uns dos outros e impressos com diversos estilos tipográficos, sugerem uma leitura não-linear e infinita, permitindo várias entradas e saídas para o leitor. Portanto, a pesquisa aborda inicialmente a evolução da ciberliteratura, a partir do paradigma do hipertexto, passando pelas poéticas digitais cíbridas, a literatura ergódica, até enfocar as manifestações literárias geradas por computador: a poesia digital, a literatura generativa e a hiperficção, conforme tipologia proposta por Pedro Barbosa. Neste percurso, são citados autores como Jay David Bolter, George Landow, Jacques Derrida, Lev Manovich, Lúcia Santaella, Giselle Bielguelman, Janet Murray, Aspen Arseth, Pedro Barbosa, Chris Funkenhouser, Rui Torres e Jorge Luiz Antônio. A dissertação prossegue através de uma apresentação da obra poética do poeta francês, e propõe uma análise comparativa entre os pontos coincidentes entre o Sintext e o Livre . Neste contexto foram destacados comentários dos autores Maurice Blanchot, Jacques Scherer, Arlindo Machado, Octavio Paz, além de outros autores já citados. O projeto culmina com uma experimentação poética em meio digital, na qual o poema Um lance de dados de Mallarmé será submetido a um motor textual semelhante ao Sintext. O processo realizado através de uma operação combinatória com os versos do poema, permitirá infinitas possibilidades recombinantes de tais versos, ampliando o sentido da obra e, simulando através das poéticas digitais, o livro, múltiplo e infinito, um dia idealizado por Mallarmé

Made available in DSpace on 2014-12-17T15:47:57Z (GMT). No. of bitstreams: 1 MarioSST_DISSERT.pdf: 2743049 bytes, checksum: da23e766aa49e7926f3ad6450145c626 (MD5) Previous issue date: 2011-02-17 / Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico / Product derivation tools are responsible for automating the development process of software product lines. The configuration knowledge, which is responsible for mapping the problem space to the solution space, plays a fundamental role on product derivation approaches. Each product derivation approach adopts different strategies and techniques to manage the existing variabilities in code assets. There is a lack of empirical studies to analyze these different approaches. This dissertation has the aim of comparing systematically automatic product derivation approaches through of the development of two different empirical studies. The studies are analyzed under two perspectives: (i) qualitative that analyzes the characteristics of approaches using specific criteria; and (ii) quantitative that quantifies specific properties of product derivation artifacts produced for the different approaches. A set of criteria and metrics are also being proposed with the aim of providing support to the qualitative and quantitative analysis. Two software product lines from the web and mobile application domains are targets of our study / Abordagens de deriva??o de produto s?o respons?veis por automatizar o processo de engenharia de aplica??o de linhas de produto de software. O conhecimento de configura??o, que ? respons?vel por relacionar o espa?o do problema com o espa?o da solu??o, desempenha um papel fundamental nas abordagens de deriva??o de produto. Cada ferramenta adota estrat?gias e t?cnicas diferentes para gerenciar o processo de deriva??o de produto e h? uma car?ncia de estudos experimentais para avaliar as diferentes abordagens. Esta disserta??o de mestrado tem como objetivo comparar sistematicamente abordagens de deriva??o autom?tica de produtos, atrav?s do desenvolvimento de estudos emp?ricos. Os estudos s?o desenvolvidos sob duas perspectivas: (i) qualitativa atrav?s da an?lise de caracter?sticas das ferramentas; e (ii) quantitativa atrav?s da quantifica??o de propriedades espec?ficas de artefatos de deriva??o produzidos para as ferramentas. Um conjunto de crit?rios e m?tricas tamb?m ? proposto com o objetivo de oferecer suporte para as an?lises qualitativas e quantitativas. Duas linhas de produto, uma para o dom?nio de sistemas web e outra para o contexto de aplica??es de dispositivos m?veis, s?o alvos do nosso estudo

Submitted by Geandra Rodrigues (geandrar@gmail.com) on 2018-03-27T13:46:07Z No. of bitstreams: 1 jonathanhunderdutragherardpinto.pdf: 6016290 bytes, checksum: 50eab93d008d20c4a60c851574b2c6f3 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2018-03-27T13:57:34Z (GMT) No. of bitstreams: 1 jonathanhunderdutragherardpinto.pdf: 6016290 bytes, checksum: 50eab93d008d20c4a60c851574b2c6f3 (MD5) / Made available in DSpace on 2018-03-27T13:57:34Z (GMT). No. of bitstreams: 1 jonathanhunderdutragherardpinto.pdf: 6016290 bytes, checksum: 50eab93d008d20c4a60c851574b2c6f3 (MD5) Previous issue date: 2018-02-19 / Este trabalho tem como contribuição o desenvolvimento de uma estratégia de equa-lização das tensões em um conversor multinível modular, como parte integrante de um sistema híbrido de armazenamento de energia. O conversor modular multinível realiza a conexão em série de módulos supercapacitores, o que possibilita aumentar a ten-são sem prejudicar a transferência rápida de energia. Em relação à outras topologias, este trabalho permite reduzir a quantidade, volume e massa do elemento magnético da estrutura do conversor. Um banco de baterias de íons de lítio também integra o sistema por intermédio de um conversor estático. Como é a fonte de maior densidade de energia, fornece a potência média requerida pelo carga. A associação com uma fonte de transferência rápida de energia permite aumentar o desempenho dinâmico, a eficiência energética e a vida útil da bateria. Com efeito, tem-se um sistema híbrido de armazenamento de energia que requer estratégias de gestão para múltiplas fontes de suprimento. Os resultados simulados considerando a estimativa da demanda de po-tência de um protótipo de veículo elétrico, são adequados e propiciam os fundamentos necessários para a construção de um protótipo. / This work is a contribution to develop a strategy equalization of tensions in a mo-dular multilevel converter as part of a hybrid system energy storage. The multilevel modular converter realizes the series connection of supercapacitor modules, which al-lows to increase the voltage without cause damages to the quick energy transfer. In relation to other topologies, it allows reduction of the quantity, volume and mass of the magnetic element of the converter structure. A lithium-ion battery bank also integrates the system via a voltage boost converter. This battery is the source of high energy density, which provides the average power required by the load. The association with a fast transfer power source allows for increased dynamic performance, energy efficiency and service life. In fact, there is a hybrid energy storage system that requires mana-gement strategies for multiple sources of supply. The simulated results were obtained considering the power demand estimation of an electric vehicle prototype.