Protecting Your Shopping Preference with Differential Privacy

Abstract

Due to numerous threats, online banks may reveal clients' buying interests. Each user may intercept their consumption amount locally before transferring it to online banks thanks to differential privacy. However, since current differential data protection solutions do not take into account resolving the noise margin issue, the straight deployment of differential data protection in online banking actually causes difficulties. In this research, we present an O-DIOR (optimal online differential private transaction) technique for setting utilisation volume caps with additional noise for online banks. We alter O-DIOR to produce a RO-DIOR schema in order to choose various boundaries that fulfil various privacy definitions. Additionally, we demonstrate that our systems can adhere to various privacy restrictions by way of a thorough theoretical study. Finally, we used our methods in tests using mobile payments to gauge their performance. The experimental findings demonstrate a considerable reduction in the relevance between consumption amount and online bank amount as well as reciprocal information privacy losses that are smaller than 0.5.