Understanding the meaning of messages exchanged be-tween software agents has long been realized as one of the key problems to realizing multi-agent systems. Forc-ing all agents to use a common vocabulary defined in a shared ontology is an oversimplified solution, especially when these agents are designed and deployed independently of each other. An alternative, and more realistic, solution would be to provide mapping services between different ontologies. In this paper, we present our work along this direction. This work combines the recently emerging semantic markup language DAML+OIL (for ontology specification), the information retrieval technique (for similarity information collection), and Bayesian reasoning (for similarity synthesis and final mapping selection), to provide ontology mapping between two classification hierarchies.