Not available / text

Not available / text

Not available / text

It is important for designers, developers, and end-users to comprehend (or explain) why a software agent acts in a particular way when situated in its operating environment. Comprehending agent behaviors in an agent-based system is a challenging task due to environmental uncertainty and the dynamics and multitude of agent interactions, which must be captured, processed, and analyzed by the human user. While traditional software comprehension answers "what is happening in the implementation?", this research takes a step further to facilitate comprehension by answering "why is the behavior happening in the implementation?". To explain agent behaviors in the implemented system, this research takes the model-checking approach for representing abstracted software behavior and the reverse engineering approach for verifying the expected behavior model against the implementation's actual behavior, while assimilating the terminology and framework from abductive reasoning. This research empirically shows that maintaining accurate background knowledge of how the implementation is expected to behave is crucial in generating accurate explanations of agent behavior. The resulting Tracing Method and accompanying Tracer Tool build on ideas from existing approaches and extend the state-of-the-art to better assist human users (of various skill levels) in comprehending agent-based software by automating many reasoning tasks. The Tracing Method is applied to two domains to demonstrate the capabilities of the Tracer Tool in (1) suggesting background knowledge updates, (2) interpreting actual behaviors from implementation executions, and (3) explaining observed agent behaviors. This research aims to help designers who want to improve agent behavior; developers who need to debug and verify agent behavior; and end-users who want to comprehend agent behaviors. / text

Mainly in business domains, the vision of gaining flexible, adaptive service environments is based on the standardization and practical proliferation of (Semantic) Web Services, ontologies, and agents. The standards of Web Services and their Service-oriented Architectures (SOA) became the standard paradigm for software component integration. Dynamic changes and the permanently increasing amount of available e-services of different domains are a challenge of Service Discovery and Composition. Mediation between different approaches and expert knowledge is often necessary for the composition of services of different domains. Semantic enhancements, Autonomic Service Discovery, and the research for more holistic concepts for the classification of e-services are current attempts of overcoming this challenge, in order to reach the ultimate goal of Autonomic SOC.

Monitoring the correct behaviour of a service-based system is a necessity and a key challenge in Service Oriented Computing. Several efforts have been directed towards the development of approaches dealing with the monitoring activity of service-based systems. However, these approaches are in general not suitable when dealing with modifications in service-based systems. Furthermore, existing monitoring approaches do not take into consideration the context of the users and how this context may affect the monitor activity. Consequently, a holistic monitor approach, capable of dealing with the dynamic nature of service-based systems and of taking into consideration the user context, would be highly desirable. In this thesis we present a monitor adaptation framework capable of dealing with changes in a service-based system and different types of users interacting with it. More specifically, the framework obtains a set of monitor rules, necessary to verify the correct behaviour of a service-based system, for a particular user. Moreover, the monitor rules verifying the behaviour of a service-based system relate to properties of the context types defined for a user. The main contributions of our work include the general characterisation of a user interacting with a service-based system and the generation of suitable monitor rules.The proposed framework can be applied to any service composition without the need of further modifications. Our work complements previous research carried on in the area of web service monitoring. More specifically, our work generates a set of suitable monitor rules - related to the user context - which are deployed in a run-time monitor component. Our framework has been tested and validated in several cases considering different scenarios.

During the late 1990s commerce and academia voiced major concerns about the problems with development processes for Web Engineering. These concerns primarily centred upon the perceived chaotic and 'ad-hoc' approach to developing Web-based applications in extremely short time-scales when compared to traditional software development. Based on personal experience, conducting a survey of current practice, and collecting supporting evidence from the literature, I proposed a set of seven criteria that need to be addressed by a successful Web engineering process: 1. Short development life-cycle times; 2. Delivery of bespoke solutions and different business models; 3. Multidisciplinary development teams; 4. Small development teams working in parallel on similar tasks; 5. Business analysis and evaluation with end-users; 6. Requirements capture and rigorous testing; 7. Maintenance (evolution) of Web-based applications. These seven criteria are discussed in detail and the relevance of each to Web engineering is justified. They are then used to provide a framework to assess the suitability of a representative sample of well-known software engineering processes for Web engineering. The software engineering processes assessed comprise: the Unified Software Development Process; Dynamic Systems Development Method; and eXtreme Programming. These seven criteria were also used to motivate the definition of the Agile Web Engineering (AWE) process. A WE is based on the principles given in the Agile Manifesto and is specifically designed to address the major issues in Web Engineering, listed above. A number of other processes for Web Engineering have been proposed and a sample of these is systematically compared against the criteria given above. The Web engineering processes assessed are: Collaborative Web Development; Crystal Orange Web; Extensions to the Rational Unified Process; and Web OPEN. In order to assess the practical application of A WE, two commercial pilot projects were carried out in a Fortune 500 financial service sector company. The first commercial pilot of A WE increased end-user task completion on a retail Internet banking application from 47% to 79%. The second commercial pilot of A WE used by an Intranet development team won the company's global technology prize for 'value add' for 2003. In order to assess the effect of AWE within the company three surveys were carried out: an initial survey to establish current development practice within the company and two further surveys, one after each of the pilot projects. Despite the success of both pilots, AWE was not officially adopted by the company for Webbased projects. My surveys showed that this was primarily because there are significant cultural hurdles and organisational inertia to adopting different process approaches for different types of software development activity within the company. If other large companies, similar to the one discussed in this dissertation, are to adopt AWE, or other processes specific to Web engineering, then many will have to change their corporate goal of a one size fits all process approach for all software technology projects.

Not available / text

The creation of a requirements model (explicitly representing functional, data, and timing requirements) typically involves accommodating viewpoints from multiple system stakeholders (e.g. multiple end-users and system maintainers). Viewpoint-oriented requirements analysis methods have been proposed by other researchers to ensure the capture of requirements imposed by all user perspectives. However, domain-modeling methodologies and CASE tools poorly address how to construct a single model given input from a variety of sources and how to maintain traceability through the synthesis process. Rather than making incremental changes to a single requirements model in response to new information, this work suggests the creation of independent models to capture the input from each viewpoint. This research provides a semi-automated method of resolving the differences between viewpoints and producing a single, traceable requirements model that embodies the merged viewpoints. Computer assistance includes the detection of consistency and completeness conflicts, the enactment of conflict resolutions, the maintenance of traceability information, and the gathering of statistics regarding the content and resolution of viewpoints. This data can provide previously unavailable insight into the progress of the requirements acquisition process and characteristics of the domain. A case study is presented to demonstrate the method and the usefulness of gathered data to software engineers and software development managers. / text

published_or_final_version / Computer Science and Information Systems / Master / Master of Philosophy