Abstract: Personalized search and vertical search are receiving more and more attention of users.User preference-based vertical search algorithm (PVSA) is proposed in this paper.By focusing on domain characteristics, PVSA uses domain topic preference vector,domain metadata weight factors,the strategy of distinguishing weights of input terms,and industry lexicon update to mine different domain preferences of different users.Experimental results show that the proposed algorithm is feasible and effective in mining users' personal preferences.