Abstract: In order to meet with the fact that all data in the real world possess the characteristic of multi-hierarchic arrangements and tense restrictions, a set of mining algorithms related to sense restriction and multi-hierarchic arrangements association rules is introduced. The core of the algorithm is to make use of minimal support and minimal confidence offered by clients, and of the statistic analysis method to find out the association rules that the clients are most concerned with. With the result of the given example, the efficiency of the algorithm is testified.