Mentioning is a social function used to explicitly point target users for specific information in online social networks(OSNs).Understanding user mention behavior can provide important insights into questions of human social behavior and improve design of social network platforms.However,most previous work mainly focus on mention network for the effect of information diffusion,few researches consider the link prediction problem of user mention behavior.In this paper,we propose an intuitive approach to predict user future mentions using link prediction method in OSNs.Specifically,we first formulate user mention prediction problem as a supervised classification task,and then extract new features including interest match,social influence and mention frequency to improve the performance of prediction.To evaluate the proposed approach,we conduct extensive experiments on Twitter dataset.The experimental results clearly show that our approach has 15%increase in precision compared with the best baseline method.