Abstract: In order to give consideration to the security and utilization of data, a privacy data publishing method based on local frequent pattern tree and proxy vector is proposed. A proxy vector based on grid environment was introduced to avoid privacy leakage. According to the data disclosure based on trusted users, the whole trajectory flow could be predicted, and the trajectory statistics of a specific region could be obtained. The local frequent pattern tree was introduced, which could effectively skip a large number of unnecessary candidate sequences, reduce the data dimension and time complexity. Experimental results show that the proposed method can ensure good security and data utilization.