Privacy-Enhancing Data Aggregation for Big Data Analytics

Data utility and data privacy are serious issues that must be considered when datasets are utilized in big data analytics such that they are traded off. That is, the datasets have high data utility and often have high risks in terms of privacy violation issues. To balance the data utility and the data privacy in datasets when they are provided to utilize in big data analytics, several privacy preservation models have been proposed, e.g., k-Anonymity, l-Diversity, t-Closeness, Anatomy, k-Likeness, and (lp1, . . . , lpn)-Privacy. Unfortunately, these privacy preservation models are highly complex data models and still have data utility issues that must be addressed. To rid these vulnerabilities of these models, a new privacy preservation model is proposed in this work. It is based on aggregate query answers that can guarantee the confidence of the range and the number of values that can be re-identified. Furthermore, we show that the proposed model is more effcient and effective in big data analytics by using extensive experiments.