Accessibility, Security, and Accuracy in Statistical Databases: The Case for the Multiplicative Fixed Data Perturbation Approach

Abstract

Organizations store data regarding their operations, employees, consumers, and suppliers in their databases. Some of the data are considered confidential, and by law, the organization is required to provide appropriate security measures in order to preserve privacy. Yet a number of companies have little or no security measures. The reason for this lack of security may, at least in part, be attributed to a lack of awareness and empirical evidence about the relative effectiveness of security mechanisms. This study investigates the effectiveness of different security mechanisms for protecting numerical database attributes. The trade-off between security, accessibility, and accuracy are examined. A comparison of different security mechanisms reveals that fixed data perturbation is preferred because it maximizes both security and accessibility. An investigation of the different approaches to fixed data perturbation indicates that multiplicative method best meets these criteria.database management, data security, data perturbation

A General Additive Data Perturbation Method for Database Security

“…Existing literature evaluates Type A, B, and D biases (Tendick 1991, Tendick and Matloff 1994, Muralidhar et al 1995, but not Type C bias. Evaluating Type C bias is important because, as stated earlier, the study of relationships between attributes can be an important part of managerial decision making.…”

“…Only Type D bias (change in the form of the distribution) will be impacted by nonnormality. Among nonnormal distributions, the issue of the appropriate method of perturbation for a log-normal data was addressed by Muralidhar et al (1995), for the univariate case. They concluded that, if a single attribute had a log-normal distribution, then MDP is better than SADP.…”

A Survey of Inference Control Methods for Privacy-Preserving Data Mining

“…Since the 80s, this has been known to be a difficult problem, subject to the tracker attack [69]. One possible strategy is to perturb the answers to queries; solutions based on perturbation can be found in [26], [54] and [76]. If perturbation is not acceptable and exact answers are needed, it may become necessary to refuse answers to certain queries; solutions based on query restriction can be found in [9] and [38].…”

Statistical Database Auditing Without Query Denial Threat

“…To control inference from aggregate statistics, many mechanisms have been proposes. They include auditing queries, e.g., Chin and Özsoyoglu (1982), Chowdhury et al (1999), query restrictions, e.g., Friedman and Hoffman (1980), Nunez et al (2007), Dobkin et al (1979), perturbation, e.g., Matloff (1986), Muralidhar et al (1999), Lee et al (2010), Muralidhar et al (1995), Li and Sarkar (2006), Sarathy et al (2002), Li and Sarkar (2013), cell suppression, e.g., Castro (2007), Fischetti and Salazar (2001), providing approximate answers, e.g., Kadane et al (2006), Garfinkel et al (2002), anonymous data collection, e.g., Kumar et al (2010), and data shuffling or swapping, e.g., Muralidhar and Sarathy (2006), Li and Sarkar (2011). A good survey of classic inference control techniques on SDBs can be found in Adam and Wortmann (1989).…”