Research on privacy preservation of dynamic datasets
BI Hong-jing · Journal of Inner Mongolia University of Science and Technology · 2010
Privacy protection for individuals has become a serious concern in many applications.Current privacy preserving data publishing techniques concentrate on the static datasets,which need onlyone-time releases.However,most of the real world data sources are dynamic,and the direct application of the existing static datasets privacy preserving techniques often causes the unexpected private information disclosure.This paper discusses various inference channels of serial releasing of dynamic datasets on medical data and proposes an efficient algorithm on the idea of invariance.The experiments show that this method protects privacy adequately and has low information loss.