Privacy-preserving multidimensional big data analytics models, methods and techniques: A comprehensive survey

The topic of privacy-preserving big data analytics is gaining momentum now, thanks to the plethora of modern application scenarios where it can be successfully applied. These range from smart city applications to e-government, from social networks to graph-like analysis tools, from logistic transport systems to business intelligence systems, and so forth. Healthcare systems, for instance, are turning out to be one of the major applicative settings where privacy-preserving big data analytics tools are extensively applied, due to the specific requirements dictated by the context (i.e., protecting individual privacy). Multidimensional big data analytics methods are a special case of big data analytics methods where the emphasis is put on the usage of multidimensional metaphors, which had a fortunate experience in OnLine Analytical Processing and Business Intelligence systems. One step forward, privacy-preserving multidimensional big data analytics, is thus of extreme interest to the research community because it combines the intrinsic privacy-preserving problem with the emerging multidimensional big data analytics methods. Inspired by these considerations, in this paper, we provide a comprehensive overview that focuses the attention on privacy-preserving multidimensional big data analytics models, methods, and techniques over big data by highlighting several characteristics of the various in-literature proposals and also providing a review of future research challenges in the investigated research domain.