The advent of smart TVs has reshaped the TV-consumer interaction by combining TVs with mobile-like applications and access to the Internet. However, consumers are still unable to seamlessly interact with the contents being streamed. An example of such limitation is TV shopping, in which a consumer makes a purchase of a product or item displayed in the current TV show. Currently, consumers can only stop the current show and attempt to find a similar item in the Web or an actual store. It would be more convenient if the consumer could interact with the TV to purchase interesting items. Towards the realization of TV shopping, this dissertation proposes a scalable multimedia content processing framework. Two main challenges in TV shopping are addressed: the efficient detection of products in the content stream, and the retrieval of similar products given a consumer-selected product. The proposed framework consists of three components. The first component performs computational and temporal aware multimedia abstraction to select a reduced number of frames that summarize the important information in the video stream. By both reducing the number of frames and taking into account the computational cost of the subsequent detection phase, this component component allows the efficient detection of products in the stream. The second component realizes the detection phase. It executes scalable product detection using multi-cue optimization. Additional information cues are formulated into an optimization problem that allows the detection of complex products, i.e., those that do not have a rigid form and can appear in various poses. After the second component identifies products in the video stream, the consumer can select an interesting one for which similar ones must be located in a product database. To this end, the third component of the framework consists of an efficient, multi-dimensional, tree-based indexing method for multimedia databases. The proposed index mechanism serves as the backbone of the search. Moreover, it is able to efficiently bridge the semantic gap and perception subjectivity issues during the retrieval process to provide more relevant results.

Dans les domaines de spécialité, les applications telles que la recherche d’information ou la traduction automatique, s’appuient sur des ressources terminologiques pour prendre en compte les termes, les relations sémantiques ou les regroupements de termes. Pour faire face au coût de la constitution de ces ressources, des méthodes automatiques ont été proposées. Parmi celles-ci, l’analyse distributionnelle s’appuie sur la redondance d’informations se trouvant dans le contexte des termes pour établir une relation. Alors que cette hypothèse est habituellement mise en oeuvre grâce à des modèles vectoriels, ceux-ci souffrent du nombre de dimensions considérable et de la dispersion des données dans la matrice des vecteurs de contexte. En corpus de spécialité, ces informations contextuelles redondantes sont d’autant plus dispersées et plus rares que les corpus ont des tailles beaucoup plus petites. De même, les termes complexes sont généralement ignorés étant donné leur faible nombre d’occurrence. Dans cette thèse, nous nous intéressons au problème de la limitation de la dispersion des données sur des corpus de spécialité et nous proposons une méthode permettant de densifier la matrice des contextes en réalisant une abstraction des contextes distributionnels. Des relations sémantiques acquises en corpus sont utilisées pour généraliser et normaliser ces contextes. Nous avons évalué la robustesse de notre méthode sur quatre corpus de tailles, de langues et de domaines différents. L’analyse des résultats montre que, tout en permettant de prendre en compte les termes complexes dans l’analyse distributionnelle, l’abstraction des contextes distributionnels permet d’obtenir des groupements sémantiques de meilleure qualité mais aussi plus cohérents et homogènes. / In specialised domains, the applications such as information retrieval for machine translation rely on terminological resources for taking into account terms or semantic relations between terms or groupings of terms. In order to face up to the cost of building these resources, automatic methods have been proposed. Among those methods, the distributional analysis uses the repeated information in the contexts of the terms to detect a relation between these terms. While this hypothesis is usually implemented with vector space models, those models suffer from a high number of dimensions and data sparsity in the matrix of contexts. In specialised corpora, this contextual information is even sparser and less frequent because of the smaller size of the corpora. Likewise, complex terms are usually ignored because of their very low number of occurrences. In this thesis, we tackle the problem of data sparsity on specialised texts. We propose a method that allows making the context matrix denser, by performing an abstraction of distributional contexts. Semantic relations acquired from corpora are used to generalise and normalise those contexts. We evaluated the method robustness on four corpora of different sizes, different languages and different domains. The analysis of the results shows that, while taking into account complex terms in distributional analysis, the abstraction of distributional contexts leads to defining semantic clusters of better quality, that are also more consistent and more homogeneous.

EB-3 est un langage de spécification développé pour la spécification des systèmes d'information. Le noyau du langage EB-3comprend des spécifications d'algèbre de processus afin de décrire le comportement des entités du système et des fonctions d'attributs qui sont des fonctions récursives dont l'évaluation se fait sur la trace d'exécution du système décrivant les attributs des entités. La vérification de propriétés temporelles en EB-3 est un sujet de grande importance pour des utilisateurs de EB-3. Dans cette thèse, on se focalise sur les propriétés de vivacité concernant des systèmes d'information exprimant l'éventualité que certaines actions puissent s'exécuter. La vérification des propriétés de vivacité se fait à l'aide de model checking. Dans un premier temps, on présente une sémantique opérationnelle deEB-3, selon laquelle les fonctions d'attributs sont évaluées pendant l'exécution du programme puis stockées. Cette sémantique nous permet de définir une traduction automatique de EB-3 vers LNT, qui est un langage simultané enrichi d'une algèbre de processus. Notre traduction assure la correspondance un à un entre les états et les transitions des systèmes étiquetés de transition correspondent respectivement à des spécifications EB-3 et LNT. Ensuite, on automatise la traduction grâce à l'outil EB3toLNT fournissant aux utilisateurs de EB-3 une tous les outils de vérification fonctionnelle disponible dans CADP. Dans le but d'améliorer les résultats de notre approche concernant le model checking, on explore des techniques d'abstraction dédiées aux systèmes d'information spécifiées en EB-3. En particulier, on se focalise sur une famille spécifique de systèmes qui s'appellent paramétriques dont le comportement varie en fonction de la valeur prédéfinie d'un paramètre du système. Enfin, on applique cette méthode dans le contexte de EB-3 / EB-3 is a specification language for information systems. The core of the EB-3 language consists of process algebraic specifications describing the behaviour of entities in a system, and attribute functions that are recursive functions evaluated on the system execution trace describing entity attributes. The verification ofEB-3 specifications against temporal properties is of great interest to users of EB-3. In this thesis, we focus on liveness properties of information systems, which express the eventuality that certain actions take place. The verification of liveness properties can beachieved with model checking. First, we present an operational semantics for EB-3 programs, in which attribute functions are computed during program evolution and their values are stored into program memory. This semantics permits us to define an automatic translation from EB-3 to LNT, a value-passing concurrent language with classical process algebra features. Our translation ensures the one-to-one correspondence between states and transitions of the labelled transition systems corresponding to theEB-3 and LNT specifications. Then, we automate this translation with the EB-3toLNT tool, thus equipping the EB-3 method with the functional verification features available in the model checking toolbox CADP. With the aim of improving the model checking results of this approach, we explore abstraction techniques for information systems specified inEB-3. In particular, we concentrate on a specific family of systems called parametric, whose behaviour is scaled in keeping with the predefined value of a system parameter. Finally, we apply this method on the EB-3 context

A Familiar Context - a survey and comparison of mathematical language in the most widely-used mathematics textbook and in the standardized national tests for grade 6 students in Sweden. Studies have shown that language has a crucial role when students learn mathematics but there is a lack of empirical surveys mapping how mathematical language is used in different practices. One aspect of mathematical language called personification, which is associated with something that generates personal interest or contributes with a familiar context, seems to affect low-performing students in a negative way in their problem solving in the subjects of algebra and geometry. The overall aim of this study is to perform an empirical survey and analysis over how the personification of the mathematical language is used in the most commonly used mathematics textbook and the standardized national tests for students in grade 6 in Sweden. To answer this purpose, two research questions have been formulated: To what extent are natural language, mathematical imagery and mathematical symbols personalized in the field of algebra and geometry? How do typical mathematical tasks differ in the subjects of algebra and geometry? Method: Selected mathematical tasks from the most commonly used mathematics textbook and the national tests were analyzed in regards to personification. These selected tasks were also analyzed in regards to the concepts, metonymy and metaphor which can be used to distinguish the core of natural language in mathematical tasks. Results: There are differences between the uses of personification in the subjects of algebra and geometry. Personification was also found to be more common in the national tests than in the mathematics textbook. A comparison between the typical tasks from each subject area in the mathematics textbook and the national tests also showed that metaphorical form, a concept like personification associated with familiar context, is used more in the national tests than in the mathematics book. Conclusions: A clear difference in the use of personification and metaphorical form between textbooks and standardized national tests in mathematics for year 6 students was identified. Such differences in language usage may impair mathematics knowledge assessment. It is therefore important to make authors aware of how language use can affect a student's problem solving ability and raises the question as to whether personification and metaphorical form should be included in testing mathematical knowledge?

No description available.

Constantemente questionado, o ensino de Matemática tem sido objeto de estudo e não faltam propostas para sugerir modificações no currículo, na metodologia, nos recursos e abordagens vigentes. É comum constatarmos dificuldades de aprendizagem relacionadas ao grau de abstração inerente à linguagem matemática. Partindo do referencial teórico estabelecido, fizemos um estudo acerca das principais competências que a Educação Básica pretende desenvolver em seus alunos, uma vez que a função principal da educação é a formação pessoal. O ENEM (Exame Nacional para o Ensino Médio) pretende avaliar o desenvolvimento de cinco competências básicas, sendo elas: expressão em diferentes linguagens, compreensão de textos e fenômenos, capacidade de argumentação ou análise, capacidade de decisão ou de síntese e capacidade de contextuação. Machado (2009) propõe o desenvolvimento da capacidade de imaginação, ou seja, da extrapolação de contextos, complementar à capacidade de contextuação, pois, segundo o autor, precisamos lidar com problemas de nossa realidade e igualmente saber resolvê-los. As abstrações matemáticas residem especialmente no polo das extrapolações, ou da imaginação. E, nesse aspecto, verificamos que há uma relação natural entre a Matemática e os contos de fadas, como também entre a Matemática e o Teatro, este último diretamente relacionado à imaginação vivida pelos atores na representação de personagens em histórias fictícias. Se, tanto o Teatro como a Matemática auxiliam no desenvolvimento da competência de abstração, nosso trabalho de pesquisa foi o de investigar a natureza e a estrutura de ambos, para estabelecer entre eles uma produtiva articulação. Neste estudo percebemos que o Teatro, originário da Grécia Antiga, surge ao lado da consciência de que todos nós representamos papéis no decorrer de nossas vidas: somos filhos, pais, funcionários, empregadores, clientes, vizinhos, religiosos, partidários etc. É própria da condição humana a necessidade da fuga (ultrapassagem) da realidade por meio das abstrações e o Teatro foi uma solução encontrada para superar o mundo das circunstâncias determinadas. O trabalho com peças teatrais nas escolas favorece o desenvolvimento da competência de abstração tão necessária à apropriação da linguagem matemática. Há também afinidade estrutural entre o Teatro e a Matemática, já que ambos possuem coerência lógica na narrativa e no desenvolvimento do algoritmo, respectivamente, que conduzem à moral da história e às resoluções de problemas, concomitantemente. Concordamos com Ortega y Gasset (2007) quando considera que o Teatro é metáfora corporificada, pois materializa as abstrações inerentes à representação teatral. Em consequência disso, acreditamos que o Teatro é uma possibilidade de materialização dos conceitos e narrativas nas aulas de Matemática. / Often questioned, the teaching of Mathematics has been studied and lots of proposals are given to suggest changes in curriculum, methodology, resources and current approaches. It is common to find learning difficulties related to the level of abstraction inherent in the mathematical language. Based on the theoretical framework established, we studied of the main competence that basic education intends to develop on students. ENEM (Exame Nacional para o Ensino Médio) intends to evaluate the development of five basic competences, which are: expression in different languages, understanding texts and phenomena, the capacity for reasoning and analysis, decision making ability or capacity of synthesis and contextualization. Machado (2009) proposes the development the capacity for imagination, that is to look beyond contexts, ability which in an additional because, according to the author, we must deal with problems of our reality and also know how to solve them. The mathematical abstractions reside especially in the center of extrapolation or imagination. In this respect we find that there is a natural connection between Mathematics and fairytales, but also between Mathematics and Theater, and Theater is related to the imagination experienced by actors in the representation of characters in fictional stories. If both the Theater and Mathematics help in the development of the mathematical abstraction competence, our research was to investigate the nature and structure of both to establish a productive articulation between them. In this research we realized that the Theater, originating in ancient Greece appears next to the consciousness that, as individuals we represent different roles throughout our lives, either as children, parents, employees, employers, customers, neighbors, religious, partisans etc. It is proper to the human condition the need of escaping (overrunning) of reality by means of abstractions and the Theater was a solution found to overcome the world of certain circumstances. The project of Theater in schools encourages the development of competence required for abstraction as the appropriation of the mathematics language. There is also structural affinity between the Theater and mathematics, as both have logical coherence in the narrative and the development of the algorithm, respectively, leading the moral and problems resolution simultaneously. We agree with Ortega y Gasset (2007) when he considers that the Theater is an embodied metaphor, materialized as the abstractions inherent in Theater. As a result we believe that Theater is a possibility of realization of the concepts and narratives in math classes.

Discrete time sequential decision processes require that an agent select an action at each time step. As humans, we plan over long time horizons and use temporal abstraction by selecting temporally extended actions such as “make lunch” or “get a masters degree”, whereby each is comprised of more granular actions. This thesis concerns itself with such hierarchical temporal abstractions in the form of macro actions and options, as they apply to goal-based Markov Decision Processes. A novel algorithm for discovering hierarchical macro actions in goal-based MDPs, as well as a novel algorithm utilizing landmark options for transfer learning in multi-task goal- based reinforcement learning settings are introduced. Theoretical properties regarding the life-long regret of an agent executing the latter algorithm are also discussed.

Consideration set formation has been suggested as an important decision-making stage prior to choice. The current research focuses on consideration sets in the memory-based choice context and addresses the gaps in the existing literature by investigating the effects of mindset abstraction on memory retrieval and the number of considered choice alternatives retrieved from memory. I propose that individuals in a concrete (vs. abstract) mindset think more contextual and specific details (vs. fewer essences) about a certain decision situation; therefore concrete and fine-grained mental representations, compared to abstract and rough representations, will activate more associated cues in memory and lead to larger memory-based consideration sets. Through a word association task, studies 1a and 1b show that concrete mindsets leads to more proliferative associations and a greater number of conceptual cues than abstract mindsets. In the domain of product consideration (i.e., snack and dinner), studies 2a and 2b directly demonstrate that individuals in concrete mindsets form a larger memory-based consideration set than ones in abstract mindsets. I further propose the Hypothesis of Top-down versus Bottom-up Approach of Memory Retrieval to explain the mechanism that underlies the mindset abstraction effect on size of memory-based consideration sets. Studies 3 and 4, using an episodic memory paradigm, support this hypothesis and reveal that the type of retrieval cues (superordinate vs. subordinate cues) used by individuals in an abstract versus a concrete mindset determines the likelihood that a brand is considered, and that the richer associations located at the subordinate level contribute to a greater number of choice alternatives that people consider in a concrete mindset. The theoretical contributions, practical implications, and future research directions of this research are finally discussed.

Nanoporous low-k dielectrics are used for integrated circuit interconnects to reduce the propagation delays, and cross talk noise between metal wires as an alternative material for SiO2. These materials, typically organosilicate glass (OSG) films, are exposed to oxygen plasmas during photoresist stripping and related processes which substantially damage the film by abstracting carbon, incorporating O and OH, eventually leading to significantly increased k values. Systematic studies have been performed to understand the oxygen plasma-induced damage mechanisms on different low-k OSG films of various porosity and pore interconnectedness. Fourier transform infrared spectroscopy, x-ray photoelectron spectroscopy and atomic force microscopy are used to understand the damage kinetics of O radicals, ultraviolet photons and charged species, and possible ways to control the carbon loss from the film. FTIR results demonstrate that O radical present in the plasma is primarily responsible for carbon abstraction and this is governed by diffusion mechanism involving interconnected film nanopores. The loss of carbon from the film can be controlled by closing the pore interconnections, He plasma pretreatment is an effective way to control the damage at longer exposure by closing the connections between the pores.

abstract: Many real-world planning problems can be modeled as Markov Decision Processes (MDPs) which provide a framework for handling uncertainty in outcomes of action executions. A solution to such a planning problem is a policy that handles possible contingencies that could arise during execution. MDP solvers typically construct policies for a problem instance without re-using information from previously solved instances. Research in generalized planning has demonstrated the utility of constructing algorithm-like plans that reuse such information. However, using such techniques in an MDP setting has not been adequately explored. This thesis presents a novel approach for learning generalized partial policies that can be used to solve problems with different object names and/or object quantities using very few example policies for learning. This approach uses abstraction for state representation, which allows the identification of patterns in solutions such as loops that are agnostic to problem-specific properties. This thesis also presents some theoretical results related to the uniqueness and succinctness of the policies computed using such a representation. The presented algorithm can be used as fast, yet greedy and incomplete method for policy computation while falling back to a complete policy search algorithm when needed. Extensive empirical evaluation on discrete MDP benchmarks shows that this approach generalizes effectively and is often able to solve problems much faster than existing state-of-art discrete MDP solvers. Finally, the practical applicability of this approach is demonstrated by incorporating it in an anytime stochastic task and motion planning framework to successfully construct free-standing tower structures using Keva planks. / Dissertation/Thesis / Masters Thesis Computer Engineering 2020