Web mining based on multi-agents

Abstract

Web mining is an important sub-branch of data mining. The data mining technology normally adopts data integration method to generate data warehouse, on which to dig the relation rules, cluster characters and get the useful module prediction and knowledge evaluation. For Web data's semi-structure and heterogeneous (mixed media) character, the Web mining technology is different to the pure mining based on database. In this paper we introduce a multi-agent cooperation module, which play as knowledge crawler, exaction machine, generalization machine and analysis machine. Through the communication between multi-agent system and agent naming server, these agents acquire useful knowledge.