Wavelet Transforms through Differential Privacy

Privacy preservation has become a major issue in many data analysis applications. When a data set is released to other parties for data analysis, privacy-preserving techniques are often required to reduce the possibility of identifying sensitive information about individuals. However, many solutions exist for privacy preserving data; differential privacy has emerged as a new paradigm for privacy protection with very conservative assumptions and guarantees the strongest privacy. In particular, for a count query answered by output data set, the noise in the result makes it vain as the result set could be equivalent to the number of Tuples in the data. This paper proposes a data publishing technique that not only ensures differential privacy, but also provides accurate results for all range-count queries, i.e., count queries where the predicate on each attribute is a range. The main aim of the solution provides a frame work called privelet that applies wavelet transforms s on the data before adding noise to it. This paper also outlines the instantiations of the Privelet for both ordinal and nominal data and theoretical Analysis is provided to prove the privacy guarantee of Privelet with the nominal wavelet transform.