根据高维数据具有方向性的特征,结合概率模糊聚类算法与粗糙集理论提出了一种粗糙的方向性模糊聚类算法。该算法在概率模糊聚类算法中引入了数据方向相似性函数，能对不确定数据进行处理。在算法中利用粗糙集中的下近似集与边界集来确定目标对象函数，属于下近似集的数据在聚类时是确定的，属于边界的数据具有模糊性。实验结果表明，该算法能有效地对高维的方向性数据进行聚类。

A probabilistic fuzzy clustering algorithm with the directional characteristics in higherdimension datasets ,which is termed as directional fuzzy clustering algorithm with rough sets(RDFC),is proposed in this paper to effectively deal with uncertainly data, RDFC leads into the data direction similarity function in probabilistic fuzzy clustering. RFDC gets object function with lower approximation sets and boundary region of rough sets, it is certain to lower approximation data, and it is fuzzy to boundary region data .The experiment results demonstrate its validity over directional higherdimension data clustering.