This paper describes the application of visualisation to the validation of software system requirements. It enhances the popular validation technique of prototyping because the execution of prototypes offer minimalistic or rudimentary presentation styles with little or no regard to customer comprehension. Visualisation enables the results generated by prototype execution to be presented using a rich repertoire of representations and more importantly, using imagery that is borrowed directly from the customer's own domain. This paper presents some of the pertinent issues in applying visualisation to this area, and describes the results of our efforts in developing an approach that visualises prototype execution behaviour.

One of the major issues in dealing with exceptions in open multi-agent systems (MAS) is lack of uniform representation of exceptions and their shared semantics. In the absence of a uniform framework different business organization may use different representations for the same exception or may interpret the same exception in different way. In order to address this issue we present an ontological approach to provide a uniform way of representing and interpreting exceptions in cross-organizational settings. This helps agents from different organizations to interpret exceptional situations in an unambiguous way and exchange exception related information using standard structures. We believe that exception ontology along with domain ontology increases the open MAS reliability and also enhances its fault tolerance capability.

The volume and variety of the Internet information is exponentially grows and therefore causes difficulties for a user to obtain information that accurately matches of the user interested. Several combination techniques are used to achieve the precise goal. This is due, firstly, to the fact that users often do not present queries to information retrieval systems that optimally represent the information they want, and secondly, the measure of a document's relevance is highly subjective and variable between different users. We address this problem with an approach that relies on evolutionary user-modelling, in order to retrieve domain-specific information. We describe an adaptive information retrieval system that learns user needs from user-provided relevance feedback. The method combines qualitative feedback measures using fuzzy inference, and quantitative feedback using genetic algorithms (GA) fitness measures. We utilise the multiagent design approach for designing an information retrieval system (IRS). The system consists of following combination of complex processes: document indexing, learning strategic for relevant feedback and user modelling using genetic algorithm, filtering and ranking the retrieve documents based on the user model. We show the design of the IRS consists of several agents that cooperate with each other and may perform in parallel to achieve the system goal.

We present the development of a Web portal to support online shopping, using fuzzy logic to enable adaptability in order to improve users' shopping experience. The adaptive interface is designed at the level of presentation of product category, through capturing and inferring dynamic user behaviour. Personalised information such as past shopping records and current shopping list is maintained. Evaluation regarding functionality and usability is performed and the results, in general, show that using a fuzzy approach for adaptation can viably be incorporated into Web interfaces.

This paper reports on ongoing research work to facilitate the user validation process based on executable formal specifications. It embodies a visualisation system which allows software developers to choose an appropriate representation of objects used in a formal specification and create a dynamic and/or static animation of these objects in an interactive and iterative fashion. This paper outlines our approach and illustrates its application in a real-world setting.

This article describes the template framework, a conceptual abstraction that enables both system and user properties to be combined. The intended purpose of the framework is illustrated through an example of applying it in the assessment of a commercially available interface for visualising and managing a directory service. The template framework is one technique for promoting the recognition of human factors within formal system modelling. The framework encourages the recognition and formalisation of system features that are relevant to effective use. Through examining interaction and interface designs in this way, assumptions about intended use, system design and user tasks can be made explicit. An informal description along with a partial formal development of the directory service and its interface is given, as well as an outline of the sort of tasks it may be expected to support. An analysis of the interface in terms of the adherence between its 'system provided view' and the 'user expected view' is carried out which is investigated in terms of a generic notion within the template model known as output correctness. The investigation focuses on determining the specification constraints required to satisfy output correctness and explores the implications of these constraints upon interface design.

The strategy of retrofitting quality onto a prototype and evolving it into a production quality system has at best led to severe problems and at worst to a complete failure. The present work reports on an environment and methodology that supports the development of high quality prototypes that can easily accommodate changes so that the prototypes can be evolved into production quality systems. An outline of the environment incorporating an architectural mechanism and associated tools that accommodates the intertwining of algebraic specifications, known as ADTSPEC, and implementations in a modular programming language, PARADOX PASCAL, is presented. A framework for applying the environment in the context of evolutionary development is detailed. An illustrative application is also presented. It demonstrates the effectiveness of the work, to support user participation at each stage of evolution and the engineering of high quality prototypes constructed in a systematic manner, incorporating the degree of rigour that is appropriate to the clarity of requirements.

AI developments have been based on an exploratory approach whereas software engineering approaches can be characterised as specification-based. We present an integrated toolset that accommodates and encourages a mixed-mode development, thereby maximising the strengths and minimising the weaknesses of both approaches. Moreover, the interactive nature of the toolset provides an effective means to validate the adequacy of the developing system.

ZAL (Z Animation in LISP) is a package which allows a Z specification to be incrementally designed and validated in an interactive environment built on Common LISP. A Z specification is transformed into an equivalent ZAL program which can be executed in order to demonstrate the functionality of the intended implementation. ZED is a full screen editor and syntax analyser for Z. Using these two tools as an integrated package encourages and facilitates an exploratory approach to formal specification which reflects the way in which many users of Z (particularly beginners) prefer to approach the specification task. The package has also been used to support the teaching of students following courses in discrete mathematics, formal specification and software development.>

A classification of requirements engineering methods based on the social assumptions made about organizations is presented. The scheme, based on a taxonomy of social theories after J. Lyotard (1984), provides a set of categories corresponding to metanarratives used within requirements engineering. A short description of each metanarrative is provided in terms of the assumptions made about the nature of society and the expected role of the requirements engineer. A decision aid for practitioners is included in the form of a table that classifies a representative sample of requirements engineering methods.>

Highlights are presented of the carry-over effects in changing from a procedural to a declarative approach. The results of a case study of programming in Prolog for a relatively simple problem are reported. The different methods that the subjects used to solve the problem are described. It is argued that they can be explained on the basis of strategies used for problem decomposition and the choice of data representation. It is also argued that the methods of solution used suffer from a carry-over effect based on a procedural approach and, in particular, that the choice of data representation used appears to be more important than the paradigm used.>

The authors briefly review three current cognitive models and report on the findings of an experiment involving the collection and analysis of video protocols of five experienced programmers. It is shown that, although there is a significant degree of similarity in the explanation offered by the three models, they focus attention on different aspects of the program design task. Furthermore, the protocols gathered provide supportive empirical evidence for these models. It is also argued that the results of these works need to be consolidated into an explanatory framework. Moreover, the orientation of this framework needs to be predictive rather than descriptive so that the necessary guidelines for software development can be made.>