Since organizational tacit knowledge resides in the owner’s brain, finding relevant experts in the specific fields is crucial to utilize this type of knowledge. In this paper, an approach to expert finding is proposed to assist the user to find the required experts. The method adopts the multi-granularity two-tuple linguistic information to construct the expert profile, that is, to model expert’s expertise. The user query the expert based on the fuzzy linguistic information. Then, the relevant experts are ranked according to the matching degree between the expert profile and the query. Finally, a numerical example is given to illustrate the effectiveness of the proposed approach.