eM~2: An Efficient Member Migration Algorithm for Ensuring k-Anonymity and Mitigating Information Loss

摘要

Privacy preservation (PP) has become an important issue in the information age to prevent expositions and abuses of personal information. This has attracted much research and k-anonymity is a well-known and promising model invented for PP. Based on the k-anonymity model, this paper introduces a novel and efficient member migration algorithm, called eM~2, to ensure k-anonymity and avoid information loss as much as possible, which is the crucial weakness of the model. In eM~2, we do not use the existing generalization and suppression technique. Instead we propose a member migration technique that inherits advantages and avoids disadvantages of existing k-anonymity-based techniques. Experimental results with real-world datasets show that eM~2 is superior to other k-anonymity algorithms by an order of magnitude.