Exploiting a company's knowledge : the adaptive search agent YASE

This paper introduces YASE, a domain-aware search agent with learning capabilities. Initially built for the research community of Roche Penzberg, YASE proved to be superior to standard search engines in the company environment due to the introduction of some simple principles: personalized ranking based on a user’s role and organizational embedding, automatic classification of documents by using domain knowledge and learning from search history. While the benefits of the learning feature need more time to be fully realized, the other two principles have proved to be surprisingly powerful. -- search agent / engine ; metadata ; personalization ; information retrieval ; YASE