Double Privacy Layer Architecture for Big Data Framework

Abstract

Big data is an emerging and very considerable technology for gathering and analyzing a huge volume of real-time produced data efficiently and effectively, but it has also a great volume of sensitive data arising invasion of privacy. Big data analyses can give us very customized and effective analysis results, but this technology can be abused for privacy invasion of personal users. This paper introduces sensitive information which can be collected and/or synthesized at the data collection stage, data analysis stage, and presentation service stage. And then this paper proposes a double layered architecture for preserving user privacy. The proposed architecture provides two steps for masking sensitive information from the big data processing databases, and these steps are presented in detail with some examples in this paper.

A Survey of System Architectures, Privacy Preservation, and Main Research Challenges on Location-Based Services

“…The following are the most prominent extensions in this category, not to be discussed in depth here, and interested readers can refer to the literatures. These are k-anonymity [54], Tessellation [55], l-diversity [56], micro-aggregation [57], data aggregation [58], t-closeness [59], and historical k-anonymity [60].…”

A Systematic Review of Security Mechanisms for Big Data in Health and New Alternatives for Hospitals

“…The first-mentioned is in charge of searching for and eliminating sensitive personal information from the data gathered, which is done in order to make the information anonymous and thus make it more difficult to identify the person in particular. The second, postfiltering layer, disguises the summarized sensitive information following big data analysis [31]. define a series of steps in which the validity of the data stored is verified externally.…”

“…By doing this we obtain a certain anonymity regarding the data, which is an important factor in the area of health [40]. As an alternative, the scheme by Cho et al (2016) can be used, which involves splitting the data into sensitive and nonsensitive but for the same purpose, that is, to make the information anonymous. The next step will be to use the proposal made by Jing (2014), which describes the use of double encryption in order to protect data [18].…”

“…On the left side, the technique by Cho et al (2016) is utilized that splits the text into sensitive and nonsensitive information that represents personal details and in general labels such as "name," respectively. On the other right side, the technique given by Subashini and Kavitha (2012) is used which fragments the text when it is required and stops until it gives value [40].…”

“…In both cases, we maintain the images without either varying or fragmenting owing to the fact that we have at our disposal the algorithm developed by Cho et al (2016) in order to ensure their efficient encryption and that no fragment of the image can be obtained without the suitable key [31]. All this information that has already been encrypted will be passed via Internet to the cloud servers.…”