A Variation-Based Genetic Algorithm for Privacy-Preserving Data Publishing
Meshari Alkhathami, Yong-Feng Ge, Hua Wang · 2024
Facing the escalating need for data sharing against the backdrop of stringent privacy demands, this study shows Variation-Based Genetic Algorithm (VB-GA) aimed at navigating the complexities of balancing privacy and utility in privacy-preserving data publishing. By integrating attribute generalisation and record suppression, VB-GA proposes a refined strategy for anonymization. It further incorporates specially designed crossover and mutation operators to optimise information exchange and enhance the anonymization process. Comprehensive experimental analysis reveals VB-GA’s ability to improve solution accuracy and hasten convergence, underscoring its efficiency in balancing privacy preservation with data utility. This research represents a pivotal advancement in data management practices amidst growing privacy valuation, highlighting genetic algorithms’ potential to refine privacy-preserving data publishing and ensure a judicious data governance approach. It delineates a pathway towards reconciling the intricate challenge of safeguarding data privacy without undermining its inherent value, setting a new standard for addressing privacy concerns in the digital age.