Martin D. Beer

Negotiation in multi-agent systems

In systems composed of multiple autonomous agents, negotiation is a key form of interaction that enables groups of agents to arrive at a mutual agreement regarding some belief, goal or plan, for example. Particularly because the agents are autonomous and cannot be assumed to be benevolent, agents must influence others to convince them to act in certain ways, and negotiation is thus critical for managing such inter-agent dependencies. The process of negotiation may be of many different forms, such as auctions, protocols in the style of the contract net, and argumentation, but it is unclear just how sophisticated the agents or the protocols for interaction must be for successful negotiation in different contexts. All these issues were raised in the panel session on negotiation.

KRAFT: knowledge fusion from distributed databases and knowledge bases

The KRAFT project aims to investigate how a distributed architecture can support the transformation and reuse of a particular class of knowledge, namely constraints, and to fuse this knowledge so as to gain added value, by using it for constraint solving or data retrieval.

Resolving Ontological Heterogeneity in the KRAFT Project

KRAFT is an agent architecture for the integration of heterogeneous information systems. The focus in KRAFT is on the integration of knowledge in the form of constraints. In this article we describe the architecture from an ontological perspective. We start by introducing the agent architecture and illustrate its application in the telecommunication-network design. We then describe how we assess the ontological heterogeneity in the domain, which problems the integration of constraint knowledge pose, and how we construct a shared ontology. Also, we describe the mapping, functions that are used to translate information between the shared and the local ontologies. Finally, we look at the direction our research is taking hereafter.

Using agents to build a practical implementation of the INCA (intelligent community alarm) system

This paper describes an agent system to demonstrate the practically of the INCA (Intelligent Community Support for the Elderly) architecture. This architecture is intended to integrate a number of autonomous systems; home monitoring, community alarms, care management systems and emergency systems command and control systems using agent technology to build effective coordinated care systems. A range of different autonomous bodies provides such care, many of which have their own management information systems already in place. Since these systems do not only contain information relevant to community care, but also all the other activities of the agent, that it would be unwilling to make available to other parties, the actual management of community care has remained primarily outside the role of current systems.nThe current demonstrator has been built using ZEUS agent-building toolkit (Nwana et al (1999)) as the basis for the development of a bench-top demonstrator to show that the INCA architecture is both scalable to realistic activity levels and integrates fully and effectively with existing computer systems in the various agencies involved, without loss of autonomy and security.

The Delivery of Effective Integrated Community Care with the Aid of Agents

The INCA (Intelligent Community Support for the Elderly) architecture is based on the integration of a number of autonomous systems; home monitoring, community alarms, care management systems and emergency systems command and control systems using agent technology to build effective co-ordinated care systems.

The provision of education and training for health care professionals through the medium of the Internet

Describes a new and experimental initiative to provide Internet‐based courses to student and professional occupational therapists in four centres in the UK, Belgium, The Netherlands and Sweden. The basis of this collaborative Occupational Therapy Internet School (OTIS) is the concept of the “Virtual College”. The aim is to support and facilitate the whole range of educational activities within a remote electronic environment. A major feature of the course organisation is the adoption of a problem‐based approach in which students will collaborate internationally to propose effective intervention in given case study scenarios. Outlines the rationale for OTIS, the content and structure of the courseware, the technical specification of the system and evaluation criteria. In addition to the more conventional Web‐based learning facilities generally offered, a number of agent‐based approaches are being adopted to assist in the management of the course by ensuring the proper delivery of course materials and to assist the functioning of project groups.

Dialogue Management in a Virtual College

The development of distributed high-bandwidth data networks at the home and business level allows many new distributed information services to be developed, based on the co-operation of autonomous agents and distributed information sources. Agents collaborate by passing messages in an agent communication language, such as KQML. The validity and meaning of these messages depends critically on the context of the communication of which they form a part. The conversation class to which the communication belongs determines this context. n nThis paper describes some of the problems associated with conversation classes derived from a distance learning application. We discuss the mechanisms necessary to satisfy a number of complex scenarios within the application domain, and show how these conditions need to be specified in terms of the policies and strategies of particular agents.

Collaborative development of courseware: Part One — Examples

Abstract Approaches to courseware development vary considerably. From examination of current research and practice, illustrated by examples taken from case studies in commercial and academic collaborative courseware production, we highlight and discuss some of the common issues identified, notably quality assurance and re-use of course material. Appropriate re-use can dramatically increase the efficiency of the production of courseware and contribute to the maintenance of consistent standards. In terms of the quality of courseware, which is difficult to assess objectively other than by obtaining user / client feedback, we suggest that the degree of coordination of the skills and efforts of the courseware team is a significant factor affecting the quality of the courseware as well as the efficiency of production. A suitably designed model of coordination should permit quality assurance procedures to be integrated into actual courseware production.

Text to Hypertext and Back Again

The establishment of hypertext as a usable medium has raised problems when confronted with the transition to and from more traditional methods of writing documents. If hypertext is to become a viable and widely used methodology, the process of transferring from linear to hypertext forms and back again must be satisfactorily achieved. This paper mainly describes work investigating the hypertext to text transformation but the reverse process (text to hypertext) is also examined. Three experiments are reported, the first looking at document structures for guiding the linearisation process, the second looking at document content and semantics. The third experiment is the MUCH experiment where a semantic net is used to investigate the linearisation process. Finally methods will be described to quantify the success of the linearisation process and some conclusions on the role of semantics and structure within hyperdocuments are presented.

A microprocessor-based logical machine system

Abstract With the increasing power of microprocessors, users are demanding the ability to run multiple applications on a single microcomputer. Each of these applications may well require its own special operating environment. Traditionally, such facilities have been provided on large computers by means of virtual machine environments. This paper proposes an approach, called Logical Machine, which provides many of the benefits of virtual machine systems without the rigid hardware restrictions. A logical machine is an abstract machine with purely logical resources. It is created by a control program called the Logical Machine Monitor ( lmm ). The lmm provides the link between the real machine and the multiple logical machines that run concurrently under its control. The motivation in proposing the logical machine system is twofold: • - as a means of providing multiple operating environments simply and effectively on a single computer • - as a means to develop a highly portable operating system. An experimental system, called to lm 286, has been developed to illustrate the feasibility of the logical machine concept. This consists of an Intel 80286-8086 dual processor configuration. The concepts of logical machine systems are introduced. The design and implementation of the lm 286 system are described. An assessment is made of how well the prototype meets the original design goals.