A Multi-Agent architecture for intelligent gathering systems
Because of the evolution of the World Wide Web, the fields of intelligent agents and multi-agent systems have become a focus of interest. This intensive research has originated several frameworks, or toolkits, that are used by the researchers to develop Internet agent-based systems. Their aim is to help the designers and engineers to build complex systems based on the agent concept. This paper presents a brief description of two of those frameworks: ZEUS and SkeletonAgent. The aim of this paper is to compare them in a common domain: to search for news in several electronic newspapers. Because every one of those multi-agent toolkits has different features, the effort to design and build systems with them, as well as their behavior is expected to be different. The paper will measure the software effort to deploy the first agent and the global multi-agent system for each technology. It also studies the software reusability and the empirical performance evaluation. Key-Words: Internet Intelligent Systems, Web Agents, Intelligent Software Agents, Software Reusability, MultiAgent Frameworks Analysis.