With the rapid development of IT, more and more information/knowledge sharing and discovery activities are moved from offline to online and many online groups have been created to facilitate such activities. However, due to the information asymmetric and information overload problems, information/knowledge holders face difficulty disseminating their information/knowledge to online groups whose members are of interests. It is also difficult for groups of users to find the most related information/knowledge. Traditional individual recommendation techniques cannot solve this problem effectively because they cannot capture the preferences of a group of users. To generate recommendations for a group of users, this paper proposes a knowledge graph-enhanced group recommendation method in which knowledge graph is used to construct comprehensive profiles for groups and information/knowledge to be recommended. The proposed group recommendation method is evaluated with real- world data and the evaluation results demonstrate the effectiveness of the proposed method.