Localizing unauthorized updates in published micro-data tables through secret order-based watermarking

摘要: The study of micro-data disclosure issue has largely focused on the privacy preservation aspect, whereas integrity a published table received limited attention. Unauthorized updates to such may lead users believe in misleading data. Traditional cryptographic stamp-based approaches allow detect unauthorized using credentials issued by data owner. However, localize exact corrupted tuples would require large number stamps be stored, leading prohibitive storage requirements. In this thesis, we explore fact that must stored particular order, which no inherent meaning under relational model. We propose series algorithms for embedding watermarks through reordering tuples. embedded detect, localize, and restore with single secret key owner, additional is required. At same time, our also efficient owner or legitimate who know key. proposed are implemented evaluated experiments real