http://hdl.handle.net/10071/140 2026-04-17T06:06:15Z http://hdl.handle.net/10071/10353 Título próprio: Communicating conflict and ambiguity in requirements engineering Autoria: Albuquerque, Maria Cabral Diogo Pinto Resumo: Effective requirements engineering in the presence of imperfection remains a major research problem. There is a lack of metaphors to aid communication about such imperfections during consultation with stakeholders. The aim of this thesis research is to improve the identification, communication, and handling of ambiguity and conflict in non-functional requirements, inadvertently introduced during the RE process. The thesis proposes a new approach based in the jigsaw puzzle metaphor, which is a novel contribution in the area of visual metaphors, and as a communication mechanism to make conflict and ambiguity explicit during stakeholder consultation meetings. This metaphor is based on jigsaw puzzles, where each puzzle piece represents a requirement. When the requirement text contains ambiguities and/or conflicts with other requirements, the respective puzzle pieces almost fit together but not perfectly. The approach presents heuristics to identify the most pertinent conflicts and ambiguities to handle and thus to make explicit through the badly-fitting matches. The gamming nature of the jigsaw puzzle metaphor, the fact it presents an easy to understand and learn language, as well as the analogy with misshapen graphical visualization (the badly-fitting matches) to represent that there is a problem, and its adequacy to a creative task as RE is; altogether are key characteristics that contribute to the adequacy and success of the jigsaw puzzle metaphor when used in stakeholder consultation meetings. In fact the jigsaw puzzle metaphor used together with the proposed method for conducting the consultation meetings with the stakeholders proved successful in:  Increasing effectiveness when compared with text presentation.  Fostering team work and communication, and improving commitment of stakeholders in co-authoring of requirements and co-responsibility in ambiguity and conflict handling.  Promoting a relaxed environment to improve team cooperation and creativity. A key contribution of this thesis is its focus on separating the processing of the information about the imperfection from the issue of communicating that imperfection. Such a separation, though critical, has not been proposed to date. Descrição: Lancaster University 2013-01-01T00:00:00Z http://hdl.handle.net/10071/3004 Título próprio: Learning from multiple sources in heterogeneous groups of agents Autoria: Nunes, Luís Resumo: The field of Multiagent Systems (MAS) is concerned with software solutions composed of several autonomous elements (agents) that interact and communicate. The scenarios these techniques apply to have special characteristics that bring about new problems, but also provide new tools to develop adequate solutions. One of the research fields that is evolving in parallel and adapting to this new type of paradigm is Machine Learning (ML). Research in ML has been increasingly focused on the development of solutions that can deal with the problems posed by MAS. The contribution of ML to this field is of utmost importance since adaptability and learning are fundamental in increasing agents’ autonomy and flexibility. This work presents a study concerning the relationship between communication and learning in a certain type of MAS. We focus on problems where we have different teams of agents, solving similar problems at different locations. Each of these teams may use different learning algorithms or heuristic solutions. In the past, learning-agents used solely the environment’s feedback as a source of information for learning. MAS provide other sources of information that can increase agents’ learning capabilities. Our goal is to determine how the communication of examples and reward information can affect the learning process. The hypotheses posed in this thesis are: That communication can improve agents’ learning performance for several learning algorithms in a specific type of problems; It is possible to enhance the benefits of communication by: using of hybrid algorithms to integrate information from different sources, using heterogeneous environments and an adequate selection of information sources. These techniques are tested in three application domains: a “toy-problem” (predator-prey), a simulation with synthetic data (load-balancing) and one using real data (traffic-control). During this study several variables that influence the performance of the exchange of information during learning where identified, namely: use of batch or specific of information; online or offline integration; number of advisors; use of heuristic advisors; heterogeneity of the environment; type of algorithm used in the integration of external information. Although not exhaustive, due to the large number of possible combinations, our research tests the effects of several of these possibilities. The initial expectations pointed towards the possibility of increasing the speed of learning and the performance by exchanging information, particularly when using heterogeneous environments. It was verified that exchanging information is beneficial, in terms of speed, performance and reliability. Contrary to our expectations the environments’ heterogeneity and other tested techniques did not show the desired effects. This fact is due, mainly, to the near-optimal performance of agents in most environments where agents are allowed to communicate. Even though there is still a long path to follow in the quest for adequate solutions, this work provides, apart from the above mentioned contributions, a review of the main difficulties found during this research that may be helpful for those that follow this path in the future. The ultimate goal of this line of research is to endow agents with the capability of learning from more sources than just the environments’ feedback in a context where information is abundant. This new perspective of learning in MAS can lead to the development of new learning paradigms, specially suited for MAS, and take us one step further in the construction of autonomous and intelligent agents. 2006-01-01T00:00:00Z http://hdl.handle.net/10071/1531 Título próprio: Um Modelo de Escalonamento Multi-Agente na Empresa Estendida Autoria: Reis, Joaquim Resumo: Este trabalho descreve um modelo de escalonamento de tarefas logísticas no contexto da cadeia de fornecimento (supply chain) de produção e distribuição, numa perspectiva multi-agente de Inteligência Artificial Distribuída e assumindo o enquadramento Empresa Estendida. O escalonamento, ou scheduling, convencionalmente definido como o problema da afectação de recursos no tempo a um conjunto de tarefas a realizar de forma coordenada e tradicionalmente limitado ao contexto das actividades produtivas intra-empresa, é estendido a um contexto inter-empresa, ou inter-organizacional. O modelo desenvolvido é organizado em dois níveis: o nível físico e o nível virtual. No nível físico modelam-se aspectos interessantes para o escalonamento convencional, nomeadamente tempo, capacidade, recursos e tarefas de produção e distribuição, enquadrados no contexto da cadeia de fornecimento cooperativa. O nível virtual é um nível de decisão e contém agentes que representam as empresas participantes na cadeia. Os agentes são responsáveis por gerir o tempo e a capacidade dos recursos do nível físico, estão interligados por relações cliente-fornecedor e podem comunicar entre si. Resumidamente, no nível físico ocorrem tarefas e fluxos de produtos através dos recursos da cadeia e de, e para, o exterior da cadeia; no nível virtual ocorrem fluxos de informação para controlo e coordenação das tarefas e fluxos de produtos pelos agentes. É definido um protocolo de interacção de alto nível entre os agentes do nível virtual do modelo, apropriado para escalonamento multi-agente. É também proposto um mecanismo de coordenação específico que permite, a cada agente, localmente percepcionar restrições temporais globais rígidas de problemas de escalonamento. Este mecanismo é integrado no protocolo de interacção de alto nível e permite identificar problemas de escalonamento temporalmente super-constrangidos e orientar e limitar a actividade de re-escalonamento. Mostram-se resultados de aplicação do modelo desenvolvido, assumindo certas restrições. Os resultados são apresentados através de exemplos que demonstram a aplicação do mecanismo de coordenação para escalonamento multi-agente acima referido. 2000-01-01T00:00:00Z http://hdl.handle.net/10071/1156 Título próprio: iDRM - Interoperability Mechanisms for Open Rights Management Platforms Autoria: Serrão, Carlos Resumo: Today’s technology is raising important challenges in the Intellectual Property (IP) field in general and to Copyright in particular [Arkenbout et al., 2004]. The same technology that has made possible the access to content in a ubiquitous manner, available to everyone in a simple and fast way, is also the main responsible for the challenges affecting the digital content IP of our days [Chiariglione, 2000]. Technological solutions and legal frameworks were created to meet these new challenges. From the technological point of view, Rights Management Systems (RMS) and Copy Protection Systems (CPS) have been developed and deployed to try to cope with them. At first, they seemed to work however, their closed and non-interoperable nature and a growing number of wrong strategic business decisions, soon lead to a strong opposition. One of the strongest negative points is the lack of rights management interoperability [Geer, 2004]. The work presented on this thesis primarily addresses the RMS interoperability problems. The objective of the thesis is to present some possible mechanisms to improve the interoperability between the different existing and emerging rights management platforms [Guth, 2003a]. Several different possible directions to rights management interoperability are pointed in this thesis. One of the most important is openness. Interoperability between different rights management mechanisms can only be achieved if they are open up to a certain level. Based on this concept, an open rights management platform is designed and presented in this thesis. Also, some of the interoperability mechanisms are presented and explained. This platform makes usage of the emerging service-oriented architectures to provide a set of distributed rights management services. Rights management solutions rely heavily on the establishment of authenticated and trust environments between its different elements. While considering different RMS, the establishment of such trust environments can be somehow complex. This thesis provides a contribution to the establishment of interoperable RMS trust environments through the usage of Public-Key Infrastructure (PKI) mechanisms. Modern rights management systems have to handle with both keying material and licenses which are used mostly to define how content is governed by the system. Managing this is a complex and hard task when different rights management solutions are considered. This thesis presents and describes a generic model to handle the key and license management life cycle, that can be used to establish a global interoperable management solution between different RMS. 2009-01-14T12:49:16Z