Community discovery is an important task in mining complex networks, and has important theoretical and application value in the terrorist organization identification, protein function prediction, public opinion analysis, etc. However, the existing community mining for the dynamic network is rarely. To address the issues, a new method for detecting the overlapping communities in dynamic networks is proposed. This method bases on the Apriori algorithm, firstly using Apriori algorithm to mine the association relations between users and form a candidate communities set. Then, takes out candidate communities from the candidate communities set in support of order and calculates the communication probability between communities meanwhile merges communities whose communication probability are too high. Finally, the ultimate overlapping communities can be made out. This method provides us with a useful way to detecting the overlapping communities in dynamic networks.