Twitter has become a crucial social networking platform for users to receive and exchange information about current events. Identifying the most influential users in social networks is pivotal for the efficient propagation of information. However, traditional influencer identification models often fall short of assessing users' influence comprehensively and neglect overlapping influence, which greatly restrains their efficiency and accuracy in identifying key users. To address such issues, we propose a method for identifying key users based on correlation and minimum distance (CM-UDM). This method can evaluate users' influence based on their individual, global, local, and propagation path characteristics in social networks. Additionally, we propose a key user decision graph in order to address the problem of overlapping users' influence. We have conducted experiments using one classical dataset and three real Twitter datasets, where the performance of our proposed method in identifying highly influential users is demonstrated.