PRIVACY PRESERVING SPATIO - TEXTUAL SKYLINES BASED ON LOCATION AGGREGATION

Abstract

To achieve cost-savings, data owners outsource their spatiotextual query services to public
cloud, which, however, may bring serious privacy issues. In this paper, we define and study the problem
of privacy-preserving spatio-textual skylines in cloud environments. To address the problem, we first
transform the locations and texts of each data object and query request into vectors and encrypt the
vectors based on an vector based encryption method to protect the data privacy. Exploiting the group of
query locations to accelerate the query processing, we further present a location-aggregation-based query
request generation method. Based on encrypted aggregated query requests, we present a corresponding
query processing algorithm for privacy-preserving spatio-textual skylines. Finally, we present analysis to
show the security guarantee of the proposed methods, and conduct thorough experiments on real datasets
to show the performance of our algorithms.