Research on improved strategy of FDM algorithm

This FDM algorithm is a fast algorithm of distributed mining of association rules. However, the algorithm is designed under the condition of non shared resource; therefore the cost of the algorithm is great. Moreover, the important information at every site is exposed to other sites that is opposite to the nowadays trend of attaching importance to privacy preserving increasingly. In order to improve FDM algorithm, a novel strategy is proposed: to compute the total support count with the method of privacy preserving, meanwhile to ensure the source of every local large item-set and local support count is covered, thus to preserve the privacy of the data distributed at sites. The efficiency and effectiveness of the proposed method have been validated by the experiments based upon some public data sets.