A Parallel Processing Method for Skyline Queries with Uncertain Preferences

A parallel processing method for Skyline queries with uncertain preferences is proposed to address the performance bottlenecks encountered by traditional centralized Skyline queries when handling large-scale, high-dimensional datasets. A probabilistic model is used to model uncertain preferences, and a parallel processing strategy based on prefix constraints and multi-layer absorption techniques is designed. By dividing the dataset into multiple disjoint subsets and executing the Skyline query algorithm independently on each subset, efficient parallel processing is achieved. Experiments show that this method can effectively reduce the computation load, improve query efficiency, and provide an efficient solution for Skyline queries under uncertain preferences.