Search engines base their operation just in keywords management; some of them are ambiguous, consequently, a lot of irrelevant results are produced. To take into account the meaning of keywords, we describe the architecture of a multi-agent system in charged of information retrieval tasks in digital libraries. The main components are termed ontological agents. They are software entities that make use of ontologies to expand queries and to avoid as much ambiguity as possible of keywords. Ontologies are also used to identify relevant sources of information. The design of these agents is based on Gaia methodology.
展开▼