An intelligent dietary planning mobile system with privacy-preserving mechanism

In recent years, the rapid development of intelligent dietary planning mobile systems, combined with the cloud technology, provides more intuitive and convenient experience for users on the diet recording and suggestion. However, this mobile-cloud structure will cause the privacy issue. Unlike the past researches which mainly focus on the user interface, response speed and suggestion accuracy, we propose a privacy-preserving mechanism for this kind of system, which is based on geometric transform on data and can be applied to most hyperplane-based clustering methods, like Support Vector Machine, K-means, etc.