Privacy preserving spatio-temporal clustering on horizontally partitioned data

Abstract

Time-stamped location information is regarded as spatio-temporal data and, by its nature, such data is highly sensitive from the perspective of privacy. In this paper, we propose a privacy preserving spatio-temporal clustering method for horizontally partitioned data which, to the best of our knowledge, was not done before. Our methods are based on building the dissimilarity matrix through a series of secure multi-party trajectory comparisons managed by a third party. Our trajectory comparison protocol complies with most trajectory comparison functions and complexity analysis of our methods shows that our protocol does not introduce extra overhead when constructing dissimilarity matrix, compared to the centralized approach. This work was funded by the Information Society Technologies programme of the European Commission, Future and Emerging Technologies under IST-014915 GeoPKDD project.