京都大学 / 新制・論文博士 / 博士(理学) / 乙第13478号 / 論理博第1578号 / 新制||理||1692(附属図書館) / (主査)教授 北川 宏, 教授 吉村 一良, 教授 竹腰 清乃理 / 学位規則第4条第2項該当 / Doctor of Science / Kyoto University / DGAM

Ensembles of distributed, autonomous and heterogenous entities that are situated in an environment, interacting over both space and time, and striving to uphold some global system coherence, mission, and goal characterize a new class of systems coined Open Computational Systems (OCS). OCS are materializing as a result of various enabling Internet technologies and examples include: ubiquitous computing, proactive computing, autonomic computing, network-centric computing, and network-centric warfare. OCS require a fundamental shift in the way we think about software development. In order to address these issues we advocate a holistic approach where models and tools come together to provide a platform for the building, understanding and monitoring of software based on the notion of these type of systems. In this research project, this was investigated by adopting the generative communication paradigm, a framework for entity communication and collaboration that will allow us to construct systems with characteristics of an OCS. Model-Driven Engineering (MDE) technologies (Domain Specific Modelling Languages and Transformation Engines) were used to provision a modelling environment for the construction, visualization and transformation of systems based on the notion of OCS. An initial mechanism was then established, and a prototype built for system understanding, verification and validation.

An enquiry into the relevance and reliability of accounting information that is provided to natural shareholders in Australia. The findings provide evidence that the information needs of shareholders are not being met, and that the existing theory of auditor independence has not integrated the views of this large and significant group of shareholders.

An enquiry into the relevance and reliability of accounting information that is provided to natural shareholders in Australia. The findings provide evidence that the information needs of shareholders are not being met, and that the existing theory of auditor independence has not integrated the views of this large and significant group of shareholders.

Thesis (M.S. in Information Technology Management and M.S. in Computer Science)--Naval Postgraduate School, September 2002. / Thesis advisor(s): Alexander Bordetsky, Richard S. Coté. Michael P. Hadley received M.S. degrees in Information Technology Management and in Computer Science. James A. Wiest received a M.S. degree in Information Technology Management. Also available online.

Stuttgart, Univ., Diplomarb., 2003.

We formulate and interpret several multi-modal registration methods inthe context of a unified statistical and information theoretic framework. A unified interpretation clarifies the implicit assumptionsof each method yielding a better understanding of their relativestrengths and weaknesses. Additionally, we discuss a generativestatistical model from which we derive a novel analysis tool, the"auto-information function", as a means of assessing and exploiting thecommon spatial dependencies inherent in multi-modal imagery. Weanalytically derive useful properties of the "auto-information" aswell as verify them empirically on multi-modal imagery. Among theuseful aspects of the "auto-information function" is that it canbe computed from imaging modalities independently and it allows one todecompose the search space of registration problems.

Made available in DSpace on 2013-08-07T18:42:21Z (GMT). No. of bitstreams: 1 000388545-Texto+Completo-0.pdf: 3555230 bytes, checksum: 87b9634f4c7e4233e5b4658ee7cdd9f2 (MD5) Previous issue date: 2007 / Knowledge acquisition in software agents can be done through experience and explicit concepts’ capturing. Both acquisition approaches require knowledge management capabilities in the agents. Knowledge management approaches are usually applied to allow organizations to find, organize, and share knowledge. This work proposes a framework for software agents’ knowledge management based on a compilation of those approaches that uses ontology-based knowledge representation. It allows agents to capture knowledge to perform their tasks and to transmit this knowledge to other agents. Ontologies and inference engines are key elements for those who want to use the framework, since ontologies are used not only to represent knowledge, but also to represent agent goals and to encapsulate message content. / Toda vez que um agente de software adquire conhecimento pela experiência ou pela absorção de novos conceitos é necessário controlar quais são os conhecimentos que o agente domina, ou seja, é necessário gerir seu conhecimento. Sabendo-se que um processo de Gestão de Conhecimento serve como um controlador dos recursos de conhecimento de uma organização, auxiliando a encontrar, organizar e compartilhar o conhecimento, o presente trabalho utilizou-se das partes deste processo para desenvolver um framework para a organização do conhecimento de agentes de software, onde o conhecimento é estruturado por meio de ontologias. O framework proposto permite aos agentes capturar conhecimento para a execução de suas tarefas e também compartilhar o conhecimento disponível para que ele possa ser reusado por outros agentes do sistema. O entendimento e uso de ontologias e motores de inferência são características fundamentais em aplicações que objetivam utilizar o framework proposto, pois, além de estruturar o conhecimento disponível, ontologias são utilizadas na representação do conteúdo de mensagens e dos objetivos dos agentes.

Made available in DSpace on 2013-08-07T18:42:25Z (GMT). No. of bitstreams: 1 000418584-Texto+Completo-0.pdf: 889263 bytes, checksum: b0fc544b8e0b9a5abafd500ef19ed9cc (MD5) Previous issue date: 2009 / Each and every framework instantiation is driven by a process, regardless it is manual or automated. In a manual process, an architect or software engineer follows a sequence of random steps towards creating an instance of a specific framework. In an automated process, the architect or the software engineer takes advantage of one or more tools that follow well-defined, configured steps to guide creation of a framework instance. To assess the real benefit of such tools is the goal of this study, where empirical data will be analyzed by comparing manual to automated process, using a framework instantiation tool in an experiment. With such data, it is expected to obtain information that shows up, in the context analyzed, the real benefits in adopting a tool to automate instantiation process. With the information gathered is also possible to figure out where those tools do not add any value, allowing suggestions to improve them in order to justify their use. / Toda a instanciação de frameworks segue um processo, seja ele manual ou automatizado. Em um processo manual, um arquiteto ou desenvolvedor segue uma seqüencia de passos aleatórios com o objetivo de criar uma instância de determinado framework. Em um processo automatizado o arquiteto ou desenvolvedor utiliza uma ou mais ferramentas que sigam passos bem definidos e pré-estabelecidos que irão auxiliar na criação da instância do framework em uso. Verificar o real benefício em usar tais ferramentas é o objetivo deste trabalho, onde dados empíricos serão analisados ao se comparar o processo manual a uma ferramenta de instanciação de frameworks em um experimento. Com isso, pretende-se obter informações que demonstrem, no contexto pesquisado, os ganhos reais de se utilizar uma ferramenta para esse fim ou que as informações demonstrem onde uma ferramenta não agrega valor, permitindo assim sugerir melhorias que justifiquem o uso de tais ferramentas.

Made available in DSpace on 2013-08-07T18:42:43Z (GMT). No. of bitstreams: 1 000390099-Texto+Completo-0.pdf: 2058232 bytes, checksum: 760b8dad5457c18a4a6bc57e9ed2e2e6 (MD5) Previous issue date: 2006 / Nowadays, object-oriented frameworks are being intensively used in software development as they are able to generate entire systems in a fast way by a reuse process. This process is also called instantiation and usually it is not trivial; it must be well documented to allow the correct creation of applications. Aspect-oriented programming has introduced new possibilities for the framework development by its composition mechanisms, but it also brought a more complex reuse process, including a new composition phase besides purely instantiation of object-oriented frameworks. Therefore, if an aspect-oriented framework does not have well documented structure, hotspots and reuse steps, it becomes very difficult for the application developer to correctly reuse it. With this in mind, this work presents an approach, called AFR (Aspect-oriented Framework Reuse) that systemizes the aspect-oriented frameworks reuse. It is composed of a set of technologies: the UML-AFR notation for hotspots documentation, the RDL+Aspects description language for reuse (instantiation and composition) activities and the Reuse Tool that offers a runtime environment for the reuse process. All these technologies are integrated to help the application developer go throw the framework’s reuse process. / Frameworks orientados a objetos são muito utilizados atualmente pela sua capacidade de gerar sistemas inteiros de forma muito rápida, por um processo de reúso também chamado de instanciação. Esse processo geralmente não é trivial, sendo necessária a sua documentação para a correta criação de aplicações. A programação orientada a aspectos introduziu novas possibilidades para o desenvolvimento de frameworks devidos a seus mecanismos de composição. Apesar disso, a introdução de aspectos nos frameworks também tornou o processo de reúso mais complexo, incluindo uma nova etapa de composição além da tradicional etapa de instanciação dos frameworks orientados a objetos. Portanto, se um framework orientado a aspectos não possuir sua estrutura, seus pontos de extensão e seu processo de reúso bem documentados, será muito difícil a sua correta reutilização por parte dos desenvolvedores de aplicação. Tendo isso em mente, este trabalho tem como objetivo apresentar uma abordagem, denominada AFR (Aspect-oriented Framework Reuse), que realiza a sistematização do processo de reúso dos frameworks orientados a aspectos. Para tanto, este trabalho irá propor um conjunto de tecnologias: a notação UML-AFR para a documentação de pontos de extensão, a linguagem RDL+Aspects para descrição das atividades envolvidas no processo de reúso (tanto instanciação quanto composição) e a ferramenta Reuse Tool para execução assistida deste processo. Essas tecnologias estão integradas de modo a auxiliar o desenvolvedor de aplicação durante o reúso do framework.