Abstract: Most exist works on privacy preserving trajectory data publishing adopt the same privacy preserving standards for all trajectories, without regard to their possibly different privacy requirements. The consequence is that the data utility of released trajectory data may be greatly reduced. In order to address this issue, a(K,ε)privacy model and an algorithm IDU-K for personalized privacy preserving trajectory data publishing are presented. The key idea of IDUK is to anonymize the trajectories personally by equivalence partitioning based on greedy clustering while assuring the information loss ratio of the released trajectory data not more han a threshold ε. Experimental analysis is designed by comparing IDU-K and the traditional algorithm on the effectiveness and data utility. Experimental results show that IDU-K is effective and feasible.