As long as there have been computers, one goal has been to be able to communicate with them using natural language. It has turned out to be very hard to implement a dialog system that performs as well as a human being in an unrestricted domain, hence most dialog systems today work in small, restricted domains where the permitted dialog is fully controlled by the system. In this thesis we present two dialog systems for communicating with an autonomous agent: The first system, the WITAS RDE, focuses on constructing a simple and failsafe dialog system including a graphical user interface with multimodality features, a dialog manager, a simulator, and development infrastructures that provides the services that are needed for the development, demonstration, and validation of the dialog system. The system has been tested during an actual flight connected to an unmanned aerial vehicle. The second system, CEDERIC, is a successor of the dialog manager in the WITAS RDE. It is equipped with a built-in machine learning algorithm to be able to learn new phrases and dialogs over time using past experiences, hence the dialog is not necessarily fully controlled by the system. It also includes a discourse model to be able to keep track of the dialog history and topics, to resolve references and maintain subdialogs. CEDERIC has been evaluated through simulation tests and user tests with good results. / <p>Report code: LiU{Tek{Lic{2006:29.</p>

Cognitive radio (CR) technology introduces a revolutionary wireless communication mechanism in terminals and network segments, so that they are able to learn their environment and adapt intelligently to the most appropriate way of providing the service for the user's exact need. By supporting multi-band, mode-mode cognitive applications, the cognitive radio addresses an interactive way of managing the spectrum that harmonizes technology, market and regulation. This dissertation gives a complete story of building a cognitive radio. It goes through concept clarification, architecture definition, functional block building, system integration, and finally to the implementation of a fully-functional cognitive radio node prototype that can be directly packaged for application use. This dissertation starts with a comprehensive review of CR research from its origin to today. Several fundamental research issues are then addressed to let the reader know what makes CR a challenging and interesting research area. Then the CR system solution is introduced with the details of its hierarchical functional architecture called the Egg Model, modular software system called the cognitive engine, and the kernel machine learning mechanism called the cognition cycle. Next, this dissertation discusses the design of specific functional building blocks which incorporate environment awareness, solution making, and adaptation. These building blocks are designed to focus on the radio domain that mainly concerns the radio environment and the radio platform. Awareness of the radio environment is achieved by extracting the key environmental features and applying statistical pattern recognition methods including artificial neural networks and k-nearest neighbor clustering. Solutions for the radio behavior are made according to the recognized environment and the previous knowledge through case based reasoning, and further adapted or optimized through genetic algorithm solution search. New experiences are gained through the practice of the new solution, and thus the CR's knowledge evolves for future use; therefore, the CR's performance continues improving with this reinforcement learning approach. To deploy the solved solution in terms of the radio's parameters, a platform independent radio interface is designed. With this general radio interface, the algorithms in the cognitive engine software system can be applied to various radio hardware platforms. To support and verify designed cognitive algorithms and cognitive functionalities, a complete reconfigurable SDR platform, called the CWT2 waveform framework, is designed in this dissertation. In this waveform framework, a hierarchical configuration and control system is constructed to support flexible, real-time waveform reconfigurability. Integrating all the building blocks described above allows a complete CR node system. Based on this general CR node structure, a fully-functional Public Safety Cognitive Radio (PSCR) node is prototyped to provide the universal interoperability for public safety communications. Although the complete PSCR node software system has been packaged to an official release including installation guide and user/developer manuals, the process of building a cognitive radio from concept to a functional prototype is not the end of the CR research; on-going and future research issues are addressed in the last chapter of the dissertation. / Ph. D.

No description available.

No description available.

This research study focused on learning lessons from the experience of designing a comprehensive case-based reasoning (CBR) tool for support of complex thinking skills. Theorists have historically identified, analyzed, and classified different thinking processes and skills. Thinking skills have been increasingly emphasized in national standards, state testing, curricula, teaching and learning resources, and research agendas. Complex thinking is the core of higher-order thinking. Complex thinking is engaged when different types of thinking and action converge to resolve a real-world, ill-structured issue such as solving a problem, designing an artifact, or making a decision. By integrating reasoning, memory, and learning in a model of cognition for learning from concrete problem-solving experience, CBR can be used to engage complex thinking. In similar and different ways, CBR theory and the related theories of constructivism and constructionism promote learning from concrete, ill-structured problem-solving experience. Seven factors or characteristics, and by extension, design requirements, that should be incorporated in a comprehensive CBR tool were extracted from theory. These requirements were consistent with five theory-, research-based facilitators of learning from concrete experience. Subsequent application of the Dick, Carey, and Carey model to these design requirements generated twenty-nine specifications for design of the tool. This research study was carried out using developmental research methodology and a standard development model. The design process included front-end analysis, creating a prototype of the tool, and evaluating the prototype. / Ph. D.

This dissertation presents an empirical approach to identifying decisions for adapting cognitive radio parameters with no a priori knowledge of the environment. Cognitively inspired radios, attempt to combine observed metrics of system performance with artificial intelligence decision-making algorithms. Current architectures trend towards hybrid combinations of heuristics, such as genetic algorithms (GA) and experiential methods, such as case-based reasoning (CBR). A weakness in the GA is its reliance on limited mathematical models for estimating bit error rate, packet error rate, throughput, and signal-to-noise ratio. The CBR approach is similarly limited by its dependency on past experiences. Both methods have potential to suffer in environments not previously encountered. In contrast, the statistical methods identify performance estimation models based on exercising defined experimental designs. This represents an experiential decision-making process formed in the present rather than the past. There are three core contributions from this empirical framework: 1) it enables a new approach to decision making based on empirical estimation models of system performance, 2) it provides a systematic method for initializing cognitive engine configuration parameters, and 3) it facilitates deeper understanding of system behavior by quantifying parameter significance, and interaction effects. Ultimately, this understanding enables simplification of system models by identifying insignificant parameters. This dissertation defines an abstract framework that enables application of statistical approaches to cognitive radio systems regardless of its platform or application space. Specifically, it assesses factorial design of experiments and response surface methodology (RSM) to an over-the-air wireless radio link. Results are compared to a benchmark GA cognitive engine. The framework is then used for identifying software-defined radio initialization settings. Taguchi designs, a related statistical method, are implemented to identify initialization settings of a GA. / Ph. D.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

L'activité de conception est un processus complexe et décisif dans le cycle de vie des produits et des procédés de fabrication. Dans le contexte actuel, les chercheurs et ingénieurs de conception notent une nette augmentation de la complexité des produits et procédés, pour satisfaire au mieux l’ensemble des exigences croissantes provenant de l’ensemble des acteurs du cycle de vie (industriels et utilisateurs) mais aussi du monde normatif. La gestion des connaissances et de l’expertise métier est un atout important pour rendre plus efficace et accélérer ce processus. Les recherches actuelles sur la gestion des connaissances font émerger des méthodes et outils performants pour identifier, formaliser, exploiter et diffuser la connaissance et les expériences issues de conceptions passées en vue de produire rapidement de nouvelles solutions. Parmi les approches existantes le Raisonnement à Partir de Cas (RàPC) et la Programmation Par Contraintes (PPC) correspondent aux besoins identifiés en Génie des Procédés. A partir de l’analyse de ces deux approches, ce travail propose un couplage du RàPC et de la PPC afin de fournir un cadre méthodologique et un outil logiciel pour une aide à la conception. Le RàPC permet de capitaliser et de remémorer les expériences passées. Toutefois, la modification de la solution passée pour répondre aux exigences du nouveau problème nécessite l’ajout de nouvelles connaissances aussi appelées connaissances d’adaptation. La PPC, quant à elle, offre justement un cadre approprié pour modéliser et gérer la connaissance permettant l’obtention d’une solution à un problème mais aussi ces connaissances d’adaptation. Outre la formalisation des connaissances d’adaptation, une des difficultés réside dans l’acquisition de ces connaissances. Dans l’approche proposée, le cycle traditionnel du RàPC a été modifié de façon à créer une boucle d’interaction avec l’utilisateur. Lorsqu’un échec d’adaptation se produit, cette boucle est activée et l’expert est sollicité pour apporter les modifications nécessaires à l’obtention d’une solution appropriée. Cette correction est l’occasion d’acquérir en ligne cette nouvelle connaissance, qui sera par la suite mise à jour et ajoutée dans le système. Un cas d’étude sur la conception d’une opération unitaire de génie des procédés permet d’illustrer l’approche. / Design is a complex and crucial process within the lifecycle of products and production processes. In the current context, design engineers and researchers notice an increasing in complexity of products and processes, in order to meet all the requirements coming from all the participants(manufacturers and users alike) in the life cycle and in the normative world as well. Knowledge management is an important asset to accelerate this process and improve its efficiency. Current research on knowledge management is producing new methods and tools to identify, formalize, exploit and disseminate knowledge from past designs experiences to produce new solutions rapidly. Among existing approaches, Case-Based Reasoning (CBR) and Constraint Programming (CP) are suited to needs identified in Process Engineering. Based on the analysis of these two approaches, this work proposes a coupling of CBR and the CP to provide a methodological framework and a software tool to assist design. The CBR allows to capitalize and retrieve past experiences. However, transforming the past solution to fit the new problem requirements needs the addition of new knowledge also known as Adaptation Knowledge. CP, meanwhile, offers an appropriate framework to model and manage knowledge required to obtain an appropriate solution to a problem, but also the adaptation knowledge. In addition to the formalization of adaptation knowledge, one of the remaining major difficulties lies in knowledge acquisition. In the proposed approach, the traditional CBR cycle has been modified to create a user interaction loop. When an adaptation failure occurs, this loop is activated and the expert is asked to make the necessary changes to achieve an appropriate solution. This correction is an opportunity to acquire this new knowledge online, which will be subsequently updated and added into the system. A case study on the design of a unit operation of Process Engineering is used to illustrate the approach

Erfahrungsmanagement (EM) ist eine Spezialform des Wissensmanagements, die sich mit aufgabenbezogenem Wissen beschäftigt. Diese Arbeit entwickelt ein Rahmenwerk für Assistenzsysteme, die Menschen bei EM-Aufgaben unterstützen. Es untersucht nicht nur technische Fragen (Erfahrungswissen sammeln, strukturieren, speichern und wiederverwenden) sondern auch organisatorische (Erfahrungswissen evaluieren und pflegen) und psychosoziale Aspekte (ein EM-System integrieren, Barrieren vermeiden, den Systemeinsatz bewerten). Fallbasierte Anwendungsbeispiele für industrielle und experimentelle Szenarien zeigen, welche Prozesse wo unterstützt oder gar teilautomatisiert werden können. Sie dienen der experimentellen Evaluierug der Fragen, die ich zu Beginn jedes Anwendungskapitels formuliert habe. / Experience Management (EM) is a special form of Knowledge Management that deals with task-based knowledge. This thesis provides a framework for assistant systems that support human beings in EM tasks. It deals not only with technical issues (how to collect, structure, store, retrieve, and reuse experiential knowledge), but als with organizational issues (how to evaluate and maintain it) and psychosocial questions (how to integrate an EM system, how to avoid barriers, how to evaluate the success of the whole system). Case-based sample applications from both, industrial and experimental scenarios, show to what extend the particular EM processes can be supported or which sub-processes can even be automated. By means of experiments with these implemented samples, we evaluate the topics that are discussed at the beginning of each application chapter.

Mesmo com a evolução tecnológica em vários setores, a área de enfermagem tem tido investimentos escassos em pesquisa e desenvolvimento capazes de atender suas expectativas, principalmente no campo da inteligência artificial. As expectativas dos enfermeiros convergem à melhora de seus processos clínicos que resultará em uma maior aproximação de seus pacientes. Além disso, há dificuldade em reunir diagnósticos de enfermagem nos hospitais, onde diversos registros clínicos e procedimentos preenchidos manualmente e armazenados ainda em folhas de papel. Esta condição compromete a legibilidade dos documentos envolvidos nos processos hospitalares, e seu arquivamento torna o processo de levantamento de informações moroso, o que acaba por inviabilizar a pesquisa à qual poderia resultar em informações importantes para melhora do processo de tomada de decisões. O objetivo desta dissertação foi trazer o estado da arte em inteligência artificial focado em raciocínio baseado em casos e sua aplicação na sistematização da assistência de enfermagem. No sentido de validar o modelo levantado foi criado um protótipo para apresentar uma aplicação que pudesse auxiliar os enfermeiros em seus processos clínicos, armazenando suas experiências em uma base de casos para futuras pesquisas. O protótipo consistiu em digitalizar diagnósticos de enfermagem pediátrica, e inserção em uma base de casos, com o intuito de avaliar a eficácia do protótipo na manipulação destes casos, em uma estrutura propicia para recuperação, adaptação, indexação e comparação de casos. Esta dissertação apresenta como resultado uma ferramenta computacional para a área da saúde, empregando uma das técnicas de inteligência artificial, Raciocínio Baseados em Casos. Os resultados foram satisfatórios devido ao alto índice de aprovação nos quesitos confiabilidade, funcionalidade, usabilidade e eficiência conforme as normas ISO/ABNT de qualidade em software. / Even with the development of technology in many industries, the nursing sector has had low investment in research and development, mainly in the field of artificial intelligence. The expectations of nurses converge to improvement over his clinical procedures that will result in a closer relationship with their patients. Moreover, there is difficulty in finding nursing diagnoses in hospitals, while clinical records and procedures are completed manually and stored even on paper. This condition compromises the readability of documents involved in the admissions process, and archiving that also becomes time-consuming process of information gathering, which derail research that could result in important information for improved decision-making. The objective of this dissertation was to bring the state of the art regarding artificial intelligence focusing on Case-Based Reasoning and its application in the systematization of nursing care. In order to validate the model a prototype was set up to demonstrate an application that would assist nurses in their clinical files, storing their experiences in a case base for future research. The prototype was to scan diagnosis pediatric nursing, and insertion into a case base in order to evaluate the effectiveness of the prototype in handling those cases. It also provides a framework for recovery, adaptation, indexing, and comparison of cases. This dissertation presents results in a computational tool for health employing one of the techniques of artificial intelligence: Case-Based Reasoning. The results were satisfactory due to high rate in terms of structure reliability, functionality, usability and efficiency according to ISO / ABNT quality in software.