In a collaborative engineering environment, to avoid unnecessary misunderstandings, participants should share semantics of engineering terminologies that are sometimes differently represented. In order to share the semantics, it is necessary to automatically identify whether two terms have equivalent semantics or not. To this end, in this study, a hybrid semantic mapping method that combines heuristics-based approach and similarity-based approach is proposed. The proposed approach utilizes ontology and Bayesian network, which consists of four phases: character matching, definition matching, similarity matching, and dependency matching. To show the effectiveness of proposed approach, semantic mapping examples for the integration of computer aided design (CAD) and product data management (PDM) system are introduced.