PRIVACY PRESERVING ASSOCIATION RULE MINING?

Introduction of high end technologies used by most of the leading software companies needs a keen focus to be kept on Daas (Data Mining as a Service). The data owner transforms data to protect corporate privacy and ships it to the server, server receives mining queries and true patterns are recovered from the extracted patterns which are received from the server. A study is done on the problem of outsourcing the association rule mining task within a corporate privacy-preserving framework. An attack model based on background knowledge and scheme is devised for privacy preserving outsourced mining. The scheme introduced ensures that from at least k-1 other transformed items each transformed item is indistinguishable with respect to the attacker's background knowledge. Results show that the techniques used provide scalability, effectiveness and privacy protection.