A Multi-Agent Architecture Based on Personalized Demand of Users for Information Retrieval on the Internet

摘要

In order to meet the personalized demand of users and help users be more effective on information retrieval on the internet, this paper presents a metasearch engine architecture based on multi-Agent and a metasearch engine algorithm. In this architecture, we regard Agent as the basic components and utilize autonomy and corporation of the Agent to achieve personal information retrieval on the internet. According to practical demands of users, different select strategies of metasearch engine algorithm are designed. The experimental results show that selection strategy of the algorithm can reflect some effects by certain characteristics of the member search engines. So, this algorithm is widely applicable.