The task of developing a framework for which agents can communicate reliably and flexibly in open systems is not trivial. This thesis addresses the dichotomy between reliable communication and facilitation of the autonomy of agents to create more flexible and emergent interactions. By the introduction of adaptations to a distributed protocol language, agents benefit from the ability to communicate interaction protocols to elucidate the social norms (thus creating more reliable communication). Yet, this approach also provides the functionality for the agent to unilaterally introduce new paths for the conversation to explore unforeseen opportunities and options (thus restoring more autonomy than possible with static protocols). The foundation of this work is Lightweight Coordination Calculus (LCC). LCC is a distributed protocol language and framework in which agents coordinate their own interactions by their message passing activities. In order to ensure that adaptations to the protocols are done in a reasonable way, we examine the use of two models of communication to guide any transformations to the protocols. We describe the use of FIPA's ACL and ultimately its unsuitability for this approach as well as the more fecund task of implementing dialogue games, an model of argumentation, as dynamic protocols. The existing attempts to develop a model that can encompass the gulf between reliability and autonomy in communication have had varying degrees of success. It is the purpose and the result of the research described in this thesis to develop an alloy of the various models, by the introduction of dynamic and distributed protocols, to develop a framework stronger than its constituents. Though this is successful, the derivations of the protocols can be dificult to reconstruct. To this end, this thesis also describes a method of protocol synthesis inspired by models of human communication that can express the dialogues created by the previous approaches but also have a fully accountable path of construction. Not only does this thesis explore a unique and novel approach to agent communication, it is tested through a practical implementation.

No description available.

Service oriented computing is crucial to a large and growing number of computational undertakings. Central to its approach are the open and network-accessible services provided by many different organisations, and which in turn enable the easy creation of composite workflows. This leads to an environment containing many thousands of services, in which a programmer or automated composition system must discover and select services appropriate for the task at hand. This discovery and selection process is known as matchmaking. Prior work in the field has conceived the problem as one of sufficiently describing individual services using formal, symbolic knowledge representation languages. We review the prior work, and present arguments for why it is optimistic to assume that this approach will be adequate by itself. With these issues in mind, we examine how, by reformulating the task and giving the matchmaker a record of prior service performance, we can alleviate some of the problems. Using two formalisms—the incidence calculus and the lightweight coordination calculus—along with algorithms inspired by information retrieval techniques, we evolve a series of simple matchmaking agents that learn from experience how to select those services which performed well in the past, while making minimal demands on the service users. We extend this mechanism to the overlooked case of matchmaking in workflows using multiple services, selecting groups of services known to inter-operate well. We examine the performance of such matchmakers in possible future services environments, and discuss issues in applying such techniques in large-scale deployments.