Getting agents to help with the research

Intelligent agents are software programs that act on behalf of human users or other systems to carry out arduous information gathering and processing tasks. Thus, an agent-oriented approach is well suited for helping "a human" to find needed information on the Internet. This article describes the design of a multi-agent system and its cooperation with existing search engines. The proposed system is set up as a middle-layer agent between the user and the search engines. This system tracks the users activity and creates a pattern using association-rule mining as a text categorization technique. Depending on how well this pattern is drawn, the system provides more reliable, accurate and well-organized results for the user than if the user goes it alone. This system is in its initial implementation phase. We believe that it is a step towards a complete multi-agent based system for searching on the Internet.