Three University of California-Berkeley researchers have written a paper discussing the first “practical approach for differential privacy.” This new method, referred to as “Elastic Sensitivity,” excludes the components of tables in large data sets and big data databases that contain individual information from the other data before running the query. 

The researchers contend the method, which employs a mathematical querying system they have named “FLEX,” can be applied to all types of databases and large-scale data. The program will enable companies and researchers to conduct analytics of Big Data, while obscuring or excluding the highly sensitive data from view and the process, at a negligible cost and drain on the system. To read the paper, click here.