L-diversity based dynamic update for large time-evolving microdata

Abstract

Data anonymization techniques based on enhanced privacy principles have been the focus of intense research in the last few years. All existing methods achieving privacy principles assume implicitly that the data objects to be anonymized are given once and fixed, which makes it unsuitable for time evolving data. However, in many applications, the real world data sources are dynamic. In such dynamic environments, the current techniques may suffer from poor data quality and/or vulnerability to inference. In this paper, we investigate the problem of updating large time-evolving microdata based on the sophisticated l-diversity model, in which it requires that every group of indistinguishable records contains at least l distinct sensitive attribute values; thereby the risk of attribute disclosure is kept under 1/l. We analyze how to maintain the l-diversity against time evolving updating. The experimental results show that the updating technique is very efficient in terms of effectiveness and data quality.