Richard Hill

From concepts to agents: towards a framework for multi-agent system modelling

Whilst tools assist the various tasks required to develop a multi-agent system (MAS), yet there still remains a gap between the generation of MAS models and program code. AUML development has enabled MAS designs to be specified in detail, including the complexities of agent communication protocols, which was a shortcoming of the Unified Modelling Language (UML) standard. However, the creation of MAS designs using AUML still requires a significant amount of design expertise on the part of the designer. We describe an approach to the development of a complex healthcare information system that defines specific steps along the path to MAS implementation. In particular we explore the use of conceptual knowledge modelling techniques by means of conceptual graphs and a transactions-based architecture for model verification during requirements gathering, together with a translation to AUML for design specification, proposing a framework to extend existing AOSE methodologies.

Transaction agent modelling: from experts to concepts to multi-agent systems

Whilst the Multi-Agent System (MAS) paradigm has the potential to enable complex heterogeneous information systems to be integrated, there is a need to represent and specify the nature of qualitative conceptual transactions in order that they are adequately comprehended by a goal-directed MAS. Using the Transaction Agent Model (TrAM) approach we examine the use of Conceptual Graphs to model an extension to an existing MAS in the community healthcare domain, whereby the existing agent capabilities are augmented with a robust set of behaviours that provide emergency healthcare management. We illustrate how TrAM serves to enrichen the requirements gathering process, whilst also supporting the definition and realisation of quantitative measures for the management of qualitative transactions.

Enhancing the initial requirements capture of multi-agent systems through conceptual graphs

A key purpose of Multi-Agent Systems (MAS) is to assist humans make better decisions given the vast and disparate information that global systems such as the Web have enabled. The resulting popularity of Agent-Oriented Software Engineering (AOSE) thus demands a methodology that facilitates the development of robust, scalable MAS implementations that recognise real-world semantics. Using an exemplar in the Community Healthcare domain and Conceptual Graphs (CG), we describe an AOSE approach that elicits the hitherto hidden requirements of a system much earlier in the software development lifecycle, whilst also incorporating model-checking to ensure robustness. The resulting output is then available for translation into Agent Oriented Unified Modelling Language (AUML) and further developed using the agent development toolkit of choice.

A Transactions Pattern for Structuring Unstructured Corporate Information in Enterprise Applications

It is known that 80-85% of all corporate information remains unstructured. As such, many enterprises rely on information systems that cause them to risk transactions that are based on lack of information (errors of omission) or misleading information (errors of commission). To address this concern, the fundamental business concept of monetary transactions is extended to include qualitative business concepts. A Transaction Model (TM) is accordingly identified that provides a structure for these unstructured but vital aspects of business transactions. By highlighting how unstructured information can be integrated into transactions, the TM provides businesses with a much more balanced view of the transactions they engage in or to discover novel transactions that they might have otherwise missed. A simple example is provided that illustrates this integration and reveals a key missing element. This discovery points to a transactions pattern that can be used to ensure that all the parties (or agents) in a transaction are identified, as well as capturing unstructured and structured information into a coherent framework. In support of the TM as a pattern, more examples of its use in a variety of domains are given. A number of enterprise applications are suggested such as in multi-agent systems, document text capture, and knowledge management.

Using Agents to Promote Effective Co-Ordination in a Community Care Environment

Community Care is about providing effective and immediate support in response to both predicted and unpredicted personal needs. It is also an area that requires extensive co-operation between independent agencies, each of which needs to meet its own objectives and targets. Agent technology provides the means by which effective co-operation can take place without compromising the essential security of both the client and the agencies involved. The autonomous nature of agents means that a variety of agents can co-operate together with various local capabilities, so long as they conform to the relevant messaging requirements. The development of an agent model is first described, followed by its realisation using the ZEUS agent-building toolkit.

Improving AOSE with an enriched modelling framework

We describe an approach to the development of a complex social care system that defines specific steps along the path to MAS implementation. In particular we explore the use of conceptual knowledge modelling techniques by means of conceptual graphs and a transactions-based architecture for model verification during requirements gathering, together with a translation to AUML for design specification, and propose a rigorous framework to enrich and extend existing AOSE methodologies. The resulting output from Transaction Agent Modelling (TrAM) can then be developed further using the agent development toolkit of choice.

Deploying an agent-based architecture for the management of community care

This paper describes a project designed to show how a global multi-agent architecture can be used to support care in the community. Services in this area are notoriously difficult to coordinate with the direct and indirect involvement of a large number of independent, autonomous agencies, each of which has its own priorities and aspirations. Also, by the very nature of the services provided the caring community is highly mobile and distributed making the collection and dissemination of information extremely difficult. The Integrated Community Care (INCA) project has been designed to show how multi-agent technology can be used to address some of these issues.

A transactions framework for effective enterprise knowledge management

An enterprise’s knowledge of itself and its environment can be greatly enhanced if it can access and integrate its divergently encoded data, information and knowledge bases. A framework is proposed that is based around transactions. Transactions recognise the optimal relationships that occur through a mutually beneficial exchange in resources. Using conceptual graphs as the lucid illustration vehicle, a simple but illuminating case study is provided. It shows how transactions can transcend the myriad quantitative and qualitative aspects of an enterprise. The transaction framework thereby provides enterprises with a useful organisational memory, thereby enabling more effective enterprise knowledge management.

Building an Agent-Based Community Care Demonstrator on a Worldwide Agent Platform

This paper describes a project designed to show how a global multi-agent architecture can be used effectively to support care in the community. Services in this area are notoriously difficult to coordinate with the direct and indirect involvement of a large number of independent, autonomous agencies, each of which has its own priorities and aspirations. Also, by the very nature of the services provided the caring community is highly mobile and distributed making the collection and dissemination of information extremely difficult. The Integrated Community Care (INCA) project has been designed to show how multi-agent technology can be used effectively to address some of these issues in a timely and cost-effective manner. The flexibility provided by such architecture allows an integration of services based on a regularly updated Individual Care Plan in a way that has been very difficult to achieve in practice. This increased flexibility not only allows the better customization of the care to the current needs of the client but also provides an effective means of recovery when things go wrong

“WHy should I do This?” MAking the information systems curriculum relevant to strategic learners

Abstract Learners who ‘strategically’ engage with information systems and computing curricula are becoming more prevalent in Higher Education institutions. Increasingly they demand more prescriptive advice from teaching staff to achieve particular grades and often fail to demonstrate the learning outcomes that the curriculum originally intended to deliver. The use of Biggs’ Constructive Alignment [1] to revise the curriculum has presented new opportunities to reflect upon the learning activities that need to be undertaken, and can result in a focus on teaching process rather than how to achieve a product. This paper presents a case study of how a business process modelling module has been developed over six years to progressively align the assessment tasks with learning activities and Intended Learning Outcomes (ILO). Early indications were that the explicit declaration of ILO made the learners appear more strategic. This shift provided new challenges for the teaching staff, particularly in terms of designing appropriate summative assessment, and presenting the ILO to learners in a meaningful and justifiable way. The use of form templates is described to assist the communication of ILO, as well as supporting the thinking necessary to build a constructively aligned curriculum. Such a curriculum places more emphasis upon the dynamics of learning interactions, and should explore what learners achieve outside of the timetabled curriculum. Finally, this work is evaluated in the context of qualitative responses to open questions asked at the end of each iteration of the module.