8 Privacy-preserving data management and operations

In the previous chapter we discussed different privacy-enhancing techniques that can be utilized in data mining operations and how to implement the k-anonymity privacy model. In this chapter we’ll explore another set of privacy models that the research community has proposed to mitigate the flaws in the k-anonymity model. Toward the end of this chapter, we’ll discuss the recent evolution of data management techniques, how these privacy mechanisms can be instrumented in database systems, and what to consider when designing a privacy-enriched database management system.

8.1 A quick recap of privacy protection in data processing and mining

You’ve seen that data analysis and mining tools are intended to extract meaningful features and patterns from collected datasets, and that direct use of original data in data mining may result in unwanted data privacy violations. Hence, we use different data sanitization operations to minimize private information disclosure. To that end, our discussion in chapter 7 and this chapter covers two particular aspects, which are summarized in figure 8.1:

8.2 Privacy protection beyond k-anonymity

8.2.1 l-diversity

8.2.2 t-closeness

8.2.3 Implementing privacy models with Python

8.3 Protecting privacy by modifying the data mining output

8.3.1 Association rule hiding

8.3.2 Reducing the accuracy of data mining operations

8.3.3 Inference control in statistical databases

8.4 Privacy protection in data management systems

8.4.1 Database security and privacy: Threats and vulnerabilities

8.4.3 Attacks on database systems