Anonymizing Data via Polynomial Regression

The amount of confidential information ac- cessible through the Internet is growing con- tinuously. In this scenario, the improvement of anonymizing methods becomes crucial to avoid revealing sensible information of indi- viduals. Among several protection methods proposed, those based on the use of linear re- gressions are widely utilized. However, there is not a reason to assume that linear regres- sion is better than using more complex poly- nomial regressions. In this paper, we present PoROP-k, a family of anonymizing methods able to protect a data set using polynomial re- gressions. We show that PoROP-k not only reduces the loss of information, but it also obtains a better level of protection compared to previous proposals based on linear regres- sions.