A growing number of initiatives in recent studies have continued to pursue improvements in e-business through the use of Business Process Management (BPM) methods and tools. However there are still some limitations that need to be addressed. While BPM has been beneficial in healthcare, there are issues around context. These involve a clear need for approaches that facilitate the understanding of context when implementing healthcare processes. This thesis introduces a BPM framework and its supporting methodology for representing and modeling context for information systems design. Scenario compositions are developed as a mechanism for linking descriptive workflow information and graphical models representing many viewpoints involved in the system development and use processes. Design patterns are used as a common information platform to bridge different context representation formats. By combining scenarios and design patterns representing context, chains of relations between context, their triggering factors, and their influences on user actions can explicitly be described and modeled.

As computing becomes ubiquitous and intelligent, it is possible for systems to adapt their behavior based on information sensed from the situational context. However, determining the context space has been taken for granted in most ubiquitous applications, and so that context-adaptive systems often miss the situational factors that are most relevant to users. The mismatch between a system's computational model and users' mental model of the context may frustrate and disorient users. This thesis describes the CCM (cognitive context model)-based approach for eliciting individual cognitive views of a context-aware task and selecting an appropriate context space for context-aware computing. It captures the situational and cognitive context for each task, using a structural architecture in which individual participants use a context view to describe their situational perspective of the task. Clustering and optimization techniques are applied to analyze and integrate context views in CCM. Developers can use the optimization output to identify an appropriate context space, specify context-aware adaptation policies and resolve run-time policy conflicts. This approach simplifies the task of context elicitation, emphasizes individual variance in context-aware activity, and helps avoid user requirements misunderstanding.

As computing becomes ubiquitous and intelligent, it is possible for systems to adapt their behavior based on information sensed from the situational context. However, determining the context space has been taken for granted in most ubiquitous applications, and so that context-adaptive systems often miss the situational factors that are most relevant to users. The mismatch between a system's computational model and users' mental model of the context may frustrate and disorient users. This thesis describes the CCM (cognitive context model)-based approach for eliciting individual cognitive views of a context-aware task and selecting an appropriate context space for context-aware computing. It captures the situational and cognitive context for each task, using a structural architecture in which individual participants use a context view to describe their situational perspective of the task. Clustering and optimization techniques are applied to analyze and integrate context views in CCM. Developers can use the optimization output to identify an appropriate context space, specify context-aware adaptation policies and resolve run-time policy conflicts. This approach simplifies the task of context elicitation, emphasizes individual variance in context-aware activity, and helps avoid user requirements misunderstanding.

Made available in DSpace on 2014-06-12T15:51:14Z (GMT). No. of bitstreams: 2 arquivo2013_1.pdf: 7106085 bytes, checksum: 47ad31fd4b9b044b146cc59b0e2bc197 (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2008 / Conselho Nacional de Desenvolvimento Científico e Tecnológico / Em uma época em que os usuários precisam processar uma quantidade cada vez maior de informação e executar tarefas cada vez mais complexas em um intervalo menor de tempo, a introdução do conceito de contexto em sistemas computacionais torna-se uma necessidade. Contexto é definido como as condições interelacionadas em que alguma coisa existe ou ocorre . Contexto é o que viabiliza a identificação do que é ou não relevante em uma dada situação. Sistemas sensíveis ao contexto são aqueles que utilizam contexto para prover informações ou serviços relevantes para a execução de uma tarefa. Projetar um sistema sensível ao contexto não é trivial, uma vez que é necessário lidar com questões relacionadas a que tipo de informação considerar como contexto, como representar essas informações, como podem ser adquiridas e processadas e como projetar o uso do contexto pelo sistema. Embora existam trabalhos que tratem desafios específicos envolvidos no desenvolvimento de sistemas sensíveis ao contexto, a maioria das soluções é proprietária ou restrita a um determinado tipo de aplicação e não são facilmente replicáveis em diferentes domínios de aplicação. Além disso, um outro problema é que projetistas de software têm dificuldade em especificar o que exatamente considerar como contexto e como projetar a sua representação, gerenciamento e uso. Esta tese propõe um framework de apoio ao projeto de sistemas sensíveis ao contexto em diferentes domínios, o qual é composto por quatro elementos principais: (i) uma arquitetura genérica para sistemas sensíveis ao contexto, (ii) um metamodelo de contexto independente de domínio, que guia a modelagem de contexto em diferentes aplicações; (iii) um conjunto de perfis UML que considera a estrutura do contexto e do comportamento sensível ao contexto; e (iv) um processo que direciona a execução de atividades relacionadas à especificação do contexto e ao projeto de sistemas sensíveis ao contexto. Para investigar a viabilidade da proposta, desenvolvemos o projeto de duas aplicações em diferentes domínios. Para uma destas aplicações, foi criado um protótipo funcional, o qual foi avaliado por usuários finais

The specification of communication standards and increased availability of sensors for mobile phones and mobile systems are responsible for a significantly increasing sensor availability in populated environments. These devices are able to measure physical parameters and make this data available via communication in sensor networks. To take advantage of the so called acquiring information for public services, other parties have to be able to receive and interpret it. Locally measured datacould be seen as a means of describing user context. For a generic processing of arbitrary context data, a model for the specification ofenvironments, users, information sources and information semantics has to be defined. Such a model would, in the optimal case, enable global domain crossing context usage and hence a broader foundation for context interpretation and integration.This thesis proposes the CII-(Context Information Integration) model for the persistence and retrieval of context information in mobile, dynamically changing, environments. It discusses the terms context and context modeling under the analysis of former publications in thefield. Further-more an architecture and prototype are presented.Live and historical data are stored and accessed by the same platform and querying processor, but are treated in an optimized fashion.Optimized retrieval for closeness in n-dimensional context-spaces is supported by a dedicated method. The implementation enables self-aware,shareable agents that are able to reason or act based upon the global context,including their own. These agents can be considered as being a part of the wholecontext, being movable and executable for all context-aware applications.By applying open source technology, a gratifying implementation of CII is feasible. The document contains a thorough discussion concerning the software design and further prototype development. The use cases at the end of the document show the flexibility and extendability of the model and its implementation as a context-base for three entirely different applications. / MediaSense

[pt] Em cenários de tomada de decisão, quando um agente, humano ou máquina, necessita de mais conhecimento para decidir devido a uma lacuna de conhecimento, surge uma necessidade de informação. Os usuários podem conscientemente tomar a iniciativa de adquirir conhecimento para preencher essa lacuna através de tarefas de buscas por informação. As consultas do usuário podem ser incompletas, imprecisas e ambíguas. Isso ocorre porque parte da informação necessária está implícita ou porque o usuário não compreende totalmente o domínio ou a tarefa que motiva a busca. Esta condição está prevista nas abordagens de busca exploratória. Embora os Grafos de Conhecimento (KG) sejam reconhecidos como fontes de informação com grande potencial para integração de dados e busca exploratória, eles são incompletos por natureza. Além disso, KGs Crowdsourced, ou KGs construídos pela integração de diversas fontes de informação de qualidade variável, precisam de uma Camada de Confiança para serem eficazes no suporte a processos de tomada de decisão. A avaliação da veracidade do conhecimento depende dos contextos das alegações e das tarefas a serem realizadas ou pretendidas (propósito). Esta pesquisa tem como objetivo preparar e consultar KGs para apoiar a exploração ciente de contexto em cenários de tomada de decisão. As contribuições incluem uma arquitetura para sistemas de apoio à decisão, composta por uma Camada de Decisão, uma Camada de Confiança e uma Camada de Conhecimento que opera sob a hipótese de Mundo Aberto Dual. A Camada de Conhecimento é composta por um Grafo de Conhecimento enriquecido de Contexto (CoaKG) e uma Máquina de Consulta baseada em CoaKG. CoaKG estende um KG padrão com mapeamentos de contexto para identificar o contexto explicitamente representado e regras para inferir o contexto implícito. A máquina de Consulta baseada em CoaKG foi projetada como uma abordagem de resposta a consultas que recupera todas as respostas contextualizadas (possíveis). A Wikidata é objeto de uma Prova de Conceito para avaliar a eficácia da Camada de Conhecimento. / [en] In decision-making scenarios, an information need arises when an agent, human, or machine needs more knowledge to decide due to a knowledge gap. Users can consciously take the initiative to acquire knowledge to fill this gap through information search tasks. User queries can be incomplete, inaccurate, and ambiguous. It occurs because part of the information needed is implicit or because the user does not fully understand the domain or the task that motivates the search. This condition is foreseen within the exploratory search approaches. Although Knowledge Graphs (KG) are recognized as information sources with great potential for data integration and exploratory search, they are incomplete by nature. Besides, Crowdsourced KGs, or KGs constructed by integrating several different information sources of varying quality, need a Trust Layer to be effective. The evaluation of knowledge truthfulness depends upon the contexts of claims and tasks being carried out or intended (purpose). This research aims to prepare and query KGs to support context-aware exploration in decision-making scenarios. The contributions include a framework for Context Augmented Knowledge Graphs-based Decision Support Systems composed of a Decision Layer, a Trust Layer, and a Knowledge Layer that operates under a Dual Open World Assumption. The Knowledge Layer comprises a Context Augmented KG (CoaKG) and a CoaKG Query Engine. CoaKG contains contextual mappings to identify explicit context and rules to infer implicit context. CoaKG Query Engine is designed as a query-answering approach that retrieves all contextualized (possible answers) from the CoaKG. Wikidata is the object of a Proof of Concept to evaluate the effectiveness of the Knowledge Layer.

The latest developments of mobile devices include a variety of hardware features that allow for more rich data collection and services. Numerous sensors, Internet connectivity, low energy Bluetooth connectivity to other devices (e.g., smart watches, activity tracker, health data monitoring devices) are just some examples of hardware that helps to provide additional information that can be beneficially used for many application domains. Among others, they could be utilized in mobile learning scenarios (for data collection in science education, field trips), in mobile health scenarios (for health data collection and monitoring the health state of patients, changes in health conditions and/or detection of emergency situations), and in personalized recommender systems. This information captures the current context situation of the user that could help to make mobile applications more personalized and deliver a better user experience. Moreover, the context related information collected by the mobile device and the different applications can be enriched by using additional external information sources (e.g., Web Service APIs), which help to describe the user’s context situation in more details. The main challenge in context modeling is the lack of generalization at the core of the model, as most of the existing context models depend on particular application domains or scenarios. We tackle this challenge by conceptualizing and designing a rich generic context model. In this thesis, we present the state of the art of recent approaches used for context modeling and introduce a rich context model as an approach for modeling context in a domain-independent way. Additionally, we investigate whether context information can enhance existing mobile applications by making them sensible to the user’s current situation. We demonstrate the reusability and flexibility of the rich context model in a several case studies. The main contributions of this thesis are: (1) an overview of recent, existing research in context modeling for different application domains; (2) a theoretical foundation of the proposed approach for modeling context in a domain-independent way; (3) several case studies in different mobile application domains.

L’énergie géothermique possède un fort potentiel aux Antilles et le développement de son exploitation gagnerait à être accompagné d’une sensibilisation des populations locales, à laquelle un effort en termes de formation et d’éducation pourrait participer. Cette thèse de didactique des sciences s’inscrit dans le cadre des études portant sur les conceptions scientifiques et sur les contextualisations didactiques. Elle fait partie du projet GEOTREF, dont l’objectif est de développer les connaissances et techniques sur la géothermie aux Antilles. La première partie de la thèse décrit les conceptions d’un échantillon représentatif d’élèves à propos de la géothermie, sur trois territoires (Guadeloupe, Martinique et Dominique). Elle révèle que les conceptions sont très différentes en fonction de l’île sur laquelle elles ont été recueillies, et qu’elles sont principalement liées aux contextes géothermiques. La deuxième partie présente un dispositif de formation, basé sur les effets de contextes. Une expérimentation met en collaboration des étudiants du Québec et de Guadeloupe. Les résultats confirment la possibilité de créer des enseignements basés sur les effets de contextes, et permettent de préciser leurs modalités d’émergences et leurs caractéristiques. L’étude montre que la confrontation des étudiants à plusieurs contextes géothermiques permet l’acquisition de conceptions expertes sur le sujet, et que les effets de contextes peuvent servir d’outil à l’acquisition des connaissances. Nous préconisons de construire les enseignements relatifs à la géothermie à partir des conceptions préalables des apprenants, tout en s’appuyant sur le contexte géothermique local. / Geothermal energy has great potential in the West Indies. The exploitation of such resources requires a deep awareness of the local populations, to which an effort in terms of training and education could participate. This science education thesis takes part in the studies related to scientific conceptions and didactic contextualization. It is also part of the GEOTREF project, which aims to develop knowledge and techniques on geothermal energy in the West Indies. The first part of the thesis describes conceptions of students in three territories (Guadeloupe, Martinique and Dominica). It reveals that the conceptions in these three islands are very different, and that they are mainly related to geothermal contexts. The second part presents a pedagogical experiment based on context effects. This experiment brings together students from Quebec and from Guadeloupe. The results confirm the possibility of a teaching based on the effects of contexts, and also make it possible to specify the emergence modalities and the characteristics of these phenomena. The study shows that the confrontation of students’ conceptions, with different geothermal contexts, help with the acquisition of expert conceptions on the subject. Context effects can be used as a tool for the acquisition of knowledge. Finally, geothermal education can be built by taking into account the learners' prior conceptions, while relying on the local geothermal context.

In this thesis, we present ContextNet, a novel general object detection framework for incorporating context cues into a detection pipeline. Current deep learning methods for object detection exploit state-of-the-art image recognition networks for classifying the given region-of-interest (ROI) to predefined classes and regressing a bounding-box around it without using any information about the corresponding scene. ContextNet is based on an intuitive idea of having cues about the general scene (e.g., kitchen and library), and changes the priors about presence/absence of some object classes. We provide a general means for integrating this notion in the decision process about the given ROI by using a pretrained network on the scene recognition datasets in parallel to a pretrained network for extracting object-level features for the corresponding ROI. Using comprehensive experiments on the PASCAL VOC 2007, we demonstrate the effectiveness of our design choices, the resulting system outperforms the baseline in most object classes, and reaches 57.5 mAP (mean Average Precision) on the PASCAL VOC 2007 test set in comparison with 55.6 mAP for the baseline. / MS / The object detection problem is to find objects of interest in a given image and draw boxes around them with object labels. With the emergence of deep learning in recent years, current object detection methods use deep learning technologies. The detection process is solely based on features which are extracted from several thousand regions in the given image. We propose a novel framework for incorporating scene information in the detection process. For example, if we know the image is taken from a kitchen, the probability of seeing a cow or an airplane decreases and observation probability of plates and persons increases. Our new detection network uses this intuition to improve the detection accuracy. Using extensive experiments, we show the proposed methods outperform the baseline for almost all object types.

O codificador aritmético binário adaptativo ao contexto adotado (CABAC – Context-based Adaptive Binary Arithmetic Coding) pelo padrão H.264/AVC a partir de perfil Main é o estado-da-arte em termos de eficiência de taxa de bits. Entretanto, o CABAC ocupa 9.6% do tempo total de processamento e seu throughput é limitado pelas dependências de dados no nível de bit (LIN, 2010). Logo, atingir os requisitos de desempenho em tempo real nos níveis mais altos do padrão H.264/AVC se torna uma tarefa árdua em software, sendo necesário então, a aceleração do CABAC através de implementações em hardware. As arquiteturas de hardware encontradas na literatura para o CABAC focam no Codificador Aritmético Binário (BAE - Binary Arithmetic Encoder) enquanto que a Binarização e Modelagem de Contextos (BCM – Binarization and Context Modeling) fica em segundo plano ou nem é apresentada. O BCM e o BAE juntos constituem o CABAC. Esta dissertação descreve detalhadamente o conjunto de algoritmos que compõem o BCM do padrão H.264/AVC. Em seguida, o projeto de uma arquitetura de hardware específica para o BCM é apresentada. A solução proposta é descrita em VHDL e os resultados de síntese mostram que a arquitetura alcança desempenho suficiente, em FPGA e ASIC, para processar vídeos no nível 5 do padrão H.264/AVC. A arquitetura proposta é 13,3% mais rápida e igualmente eficiente em área que os melhores trabalhos relacionados nestes quesitos. / Context-based Adaptive Binary Arithmetic Coding (CABAC) adopted in the H.264/AVC main profile is the state-of-art in terms of bit-rate efficiency. However, CABAC takes 9.6% of the total encoding time and its throughput is limited by bit-level data dependency (LIN, 2010). Moreover, meeting real-time requirement for a pure software CABAC encoder is difficult at the highest levels of the H.264/AVC standard. Hence, speeding up the CABAC by hardware implementation is required. The CABAC hardware architectures found in the literature focus on the Binary Arithmetic Encoder (BAE), while the Binarization and Context Modeling (BCM) is a secondary issue or even absent in the literature. Integrated, the BCM and the BAE constitute the CABAC. This dissertation presents the set of algorithms that describe the BCM of the H.264/AVC standard. Then, a novel hardware architecture design for the BCM is presented. The proposed design is described in VHDL and the synthesis results show that the proposed architecture reaches sufficiently high performance in FPGA and ASIC to process videos in real-time at the level 5 of H.264/AVC standard. The proposed design is 13.3% faster than the best works in these items, while being equally efficient in area.