With the rise of social network, more and more tourists are willing to share their own travel experiences on the Internet. Due to the unique characteristics of tourism, information gathering is very important before tourists decide where the destination is. Social networking site provide a new way for consumers collecting information, but too much information would lead to information overload. If social networking sites could understand what kind of information consumers need, and help consumers to gather useful and trustworthy information quickly. It would provide the right information to consumers in the most effective ways. In the past, relevant researches focus on the influence of friends and visual group separately, however, in the social network environment, friends and visual group have already combined together and there is no specific classification now. Therefore, this study will investigate the influence of the friends on social networking sites (including real and virtual group) and experts, and then compare the influence between friends and experts. Furthermore, use the decisions of friends to predict the preferences of consumers. The result shows that: (1) the influence of friends on social networking sites is stronger than experts (2) the influence of friends on social networking sites is dependent on the EdgeRank (3) the preferences of friends can predict the preferences of consumers.
