A multi-agent framework for web information foraging: Application to MedlinePlus

Abstract

A multi-agent architecture is proposed for an intelligent system capable of foraging information from the Web in an efficient and rapid way. We distinguish three important phases. The first one is a learning process for localizing the most relevant pages that might interest the user. The second consists in an incremental learning taking into account the changes that undergoes the Web. Finally, the last step simulates the information foraging process once the authorities are learned. Experiments were conducted on MedlinePlus, a public site dedicated for research in the domain of Health. The results are promising either for those related to Web regularities and for surfing time.