A quantifying method for trade-off between privacy and utility

Many anonymization methods have been used in data publishing and data mining. In the meantime, they reduce the utility of the dataset. So it is important to consider the tradeoff between privacy and utility. Quantifying the trade-off between usefulness and privacy of dataset has been the subject of much research in recent years. In this paper, we provide the concepts of privacy loss and utility loss and also give a method to quantify them using divergence distance in probability theory. And then, we evaluate our methodology on the Adult dataset from the UCI machine learning repository. Our result shows the relationship between privacy and utility, and also provide data users how to choose the right trade-off between privacy and utility. Finally, we conclude and show the future research direction on how to select best divergence measurement.