An Adaptive Lattice of Full-Domain Generalization for Sequential Data Publishing

• Published in: 2022 Joint International Conference on Digital Arts, Media and Technology with ECTI Northern Section Conference on Electrical, Electronics, Computer and Telecommunications Engineering (ECTI DAMT & NCON) pp. 354 - 358
• Main Authors: Soontornphand, Torsak, Srisungsittisunti, Bowonsak, Mekruksavanich, Sakorn
• Format: Conference Proceeding
• Language: English
• Published: IEEE 26.01.2022
• Subjects: Data privacy; full-domain generalization; l -diversity; lattice of domain generalization; Lattices; Publishing; sequential data; Transforms
• ISSN: 2768-4644
• DOI: 10.1109/ECTIDAMTNCON53731.2022.9720350

Sequential data publishing is a scenario of multiple releases in which the attributes are changed for each release. To protect the privacy of individuals in a sequential dataset, l - diversity can be applied, and full-domain generalization is a generalization technique that can transform a given dataset to satisfy the l -diversity requirement. In full-domain generalization, an answer of domain generalization can be searched from a lattice of domain generalization. However, a lattice of domain generalization consists of all possible domain generalizations, i.e., nodes, that is not efficient for searching. Therefore, this paper presents an adaptive lattice of domain generalization for a sequential dataset that can adapt the number of nodes in the lattice based on our proposed common root node. Our experimental results show that the adaptive lattice of domain generalization can considerably reduce the number of nodes in the lattice of domain generalization compared with the traditional lattice.