Many important and useful applications for software agents require multiple agents on a network that communicate with each other. Such agents must find each other and perform a useful joint computation without having to know about every other such agent on the network. As an example, this paper describes a matchmaker system, designed to find people with similar interests and introduce them to each other. The matchmaker is designed to introduce everyone, unlike conventional Internet media which only allow those who take the time to speak in public to be known. The paper details how the agents that make up the matchmaking system can function in a decentralized fashion, yet group themselves into clusters which reflect their users ’ interests; these clusters are then used to make introductions or allow users to send messages to others who share their interests. The algorithm uses referrals from one agent to another in the same fashion that word-of-mouth is used when people are looking for an expert. Several prototypes of various parts of the system have been implemented, and the most recent results, including simulations of up to 1000 such agents, are presented.

A referral system is a multiagent system whose member agents are capable of giving and following referrals. The specific cases of interest arise where each agent has a user. The agents cooperate by giving and taking referrals so each can better help its user locate relevant information. This use of referrals mimics human interactions and can potentially lead to greater effectiveness and efficiency than in single-agent systems. Existing approaches

Many important and useful applications for software agents require multiple agents on a network that communicate with each other. Such agents must find each other and perform a useful joint computation without having to know about every other such agent on the network. This paper describes a matchmaker system, designed to find people with similar interests and introduce them to each other. The matchmaker is designed to introduce everyone, unlike conventional Internet media which only allow those who take the time to speak in public to be known. The paper details how the agents that make it up the matchmaking system can function in a decentralized fashion, yet can group themselves into clusters which reflect their users' interests; these clusters are then used to make introductions or allow users to send messages to others who share their interests. The algorithm uses referrals from one agent to another in the same fashion that word-of-mouth is used when people are looking for an exper...

Large scale open multi-agent systems where agents need services of other agents but may not know their contact information require agent location mechanisms. Solutions to this problem are usually based on middle-ware such as matchmakers, brokers, yellow-pages agents and other middle agents. The disadvantage of these is that they impose infrastructure, protocol and communication overheads, and they do not easily scale up. We suggest a new approach to agent location, which does not require middle agents and protocols for using them. Our approach is simple and scales up with no infrastructure or protocol overheads, thus may be very useful for large scale MAS. In this paper, we analytically study the properties of our approach and discuss its advantages. 1 Introduction Multi-agent systems (MAS) are taking an increasing role in the solution of highly distributed computational problems in dynamic, open domains. We assume that large-scale open MAS will be an inevitable part of this trend. T...

The use of agents of many different kinds in a variety of fields of computer science and artificial intelligence is increasing rapidly and is due, in part, to their wide applicability. The richness of the agent metaphor that leads to many different uses of the term is, however, both a strength and a weakness: its strength lies in the fact that it can be applied in very many different ways in many situations for different purposes; the weakness is that the term agent is now used so frequently that there is no commonly accepted notion of what it is that constitutes an agent. This paper addresses this issue by applying formal methods to provide a defining framework for agent systems. The Z specification language is used to provide an accessible and unified formal account of agent systems, allowing us to escape from the terminological chaos that surrounds agents. In particular, the framework precisely and unambiguously provides meanings for common concepts and terms, enables alternative models of particular classes of system to be described within it, and provides a foundation for subsequent development of increasingly more refined concepts.

. There is an increasing demand for designers and developers to construct ever larger multi-agent systems. Such systems will be composed of hundreds or even thousands of autonomous agents. Moreover, in open and dynamic environments, the number of agents in the system at any one time will uctuate signicantly. To cope with these twin issues of scalability and variable numbers, we hypothesize that multiagent systems need to be both self-building (able to determine the most appropriate organizational structure for the system by themselves at runtime) and adaptive (able to change this structure as their environment changes). To evaluate this hypothesis we have implemented such a multiagent system and have applied it to the domain of automated trading. Preliminary results supporting the rst part of this hypothesis are presented: adaption and self-organization do indeed make the system better able to cope with large numbers of agents. 1 Introduction When designing or buildin...

Recent years have seen dramatic and sustained growth in the amount of genomic data being generated, including in late 1999 the first complete sequence of a human chromosome. The challenge now faced by biological scientists is to make sense of this vast amount of accumulated and accumulating data. Fortunately, numerous databases are provided as resources containing relevant data, and there are similarly many available programs that analyse this data and attempt to understand it. However, the key problem in analysing this genomic data is how to integrate the software and primary databases in a flexible and robust way. The wide range of available programs conform to very different input, output and processing requirements, typically with little consideration given to issues of integration, and in many cases with only token efforts made in the direction of usability. In this paper, we introduce the problem domain and describe GeneWeaver, a multi-agent system for genome analys...

Multi-Agent Systems are a promising way of dealing with large complex problems. However, it is not yet clear just how much complexity or pre-existing structure individual agents must have to allow them to work together e#ectively. In this paper, we ask to what extent agents with minimal resources, local communication and without a directory service can solve a consumer-provider matchmaking problem. We are interested in finding a solution that is massively scalable and can be used with resource poor agents in an open system. We create a model involving random search and a grouping procedure. Through simulation of this model, we show that peer-to-peer communication in a environment with multiple copies of randomly distributed like clients and providers is su#cient for most agents to discover the service consumers or providers they need to complete tasks. We simulate systems with between 500 and 32,000 agents, between 10 and 2000 categories of services, and with three to six services required by each agent. We show that, for instance, in a system with 80 service categories and 2000 agents, each requiring three random services between 93% and 97% of possible matches are discovered. Such a system can work with at least 90 di#erent service categories and tens of thousands of agents.

A question frequently asked in multi-agent systems (MASs) concerns the efficient search for suitable agents to solve a specific problem. To answer this question, different types of middle agents are usually employed. The performance of middle agents relies heavily on the matchmaking algorithms used. Matchmaking is the process of finding an appropriate provider for a requester through a middle agent. There has been substantial work on matchmaking in different kinds of middle agents. To our knowledge, almost all currently used matchmaking algorithms missed one point when doing matchmaking – the matchmaking is only based on the advertised capabilities of provider agents. The actual performance of provider agents in accomplishing delegated tasks is not considered at all. This results in the inaccuracy of the matchmaking outcomes as well as the random selection of provider agents with the same advertised capabilities. The quality of service of different service provider agents varies from one agent to another even though they claimed they have the same capabilities. To this end, it is argued that the practical performance of service provider agents has a significant impact on the matchmaking outcomes of middle agents. An improvement to matchmaking algorithms is proposed, which makes the algorithms have the ability to consider the track records of agents in accomplishing delegated tasks. How to represent, accumulate, and use track records as well as how to give initial values for track records in the algorithm are discussed. A prototype is also built to verify the algorithm. Based on the improved algorithm, the matchmaking outcomes are more accurate and reasonable.

As genome projects produce increasingly large quantities of sequence data, fast and reliable sequence analysis methods are required. Basic methods for comparing pairs of sequences or detecting patterns are well-developed, and now the key problem in analysing this genomic data is how to integrate the software and primary databases in a flexible and robust way. The wide range of available programs conform to very different input, output and processing requirements, typically with little consideration given to issues of integration. Key to addressing these issues appropriately is not to consider them as a result of the biological domain, but instead as an information processing problem that suggests nothing as much as an agent-based approach. In this paper, we introduce GeneWeaver, a multi-agent system for bioinformatics, and describe in detail the agent interactions which allow the integration and management of analysis methods and data.