关键词: 概率Skyline, 分布式数据库, 不确定数据, 支配概率, 移动对象

Abstract: Skyline computation has played a significant role in the fields of multi-criteria decision making, data mining and database visualization. The uncertainty of moving objects makes the dominant relationship of data instable, which will affect global probabilistic skyline set. In this paper, the updating of continuous probabilistic Skyline queries is studied, which is under distributed environment with the uncertainty of moving objects. A continuous probabilistic Skyline queries algorithm in order to reduce communication cost called CDPS-UMO is proposed. The change of local probabilistic Skyline points in local sites is traced. The SM（Sort Method） is introduced, and the feedback rules are proposed, which will reduce the correspondence and computation cost. A base algorithm naive is proposed to be compared with CDPS-UMO. The experiments have positive results that show effectiveness of the proposed algorithm.

Key words: probabilistic Skyline, distributed database, uncertain data, dominant probability, moving objects