A new model of weighted association rule was presented in order to solve the problem that data item have not the same importance in datasets. Based on this model, a new algorithm of mining weighted association rules with multiple minimum supports was proposed. The algorithm allows the user to specify varied minimum supports and items weights to reflect the importance and frequency of each data item in datasets .The algorithm aims to deal with problem that items have different importance and varied frequency in transaction database and find more interesting rules which involve both frequent and rare items. The correlative properties of model and algorithm were given and the theories were proved. Finally, the algorithm was tested on the experimental data. Experiment results show that the new algorithm is effective for large databases.