Traditional association rules and temporal association rule mining only consider the relationship between the items in the transaction database, don't take into account that the association rules mined out may exist a certain relationship. Therefore, this paper propose the second association rule mining on top of the window-based temporal association rules. Put the temporal association rules in a window as a transaction occurs, dig hidden relationships between the temporal association rules. Finally, we provide a case study, and find the relationship hidden in the temporal association rules.