Improving privacy preserving methods to enhance data mining for correlation research

Privacy Preservation techniques have been developed to de-identify an individual from a published dataset. In consequence, full potential details of the data are not being published due to methods used today. This research proposes a more competent way to approach anonymity in a dataset. Two pieces of change should be implemented to make privacy-preserving methods more efficient: 1/2k criteria and attribute importance introduction. Together, this new way of de-identifying data could prove to be more effective to publish the highest amount of the original information.