Paper 2017/405

Security Analysis of ``PSLP: Privacy-Preserving Single-Layer Perceptron Learning for e-Healthcare''

Jingjing Wang, Xiaoyu Zhang, Jingjing guo, and Jianfeng Wang

Abstract

With the synchronous development of both cloud computing and machine learning techniques, the clients are preferring to resort to the cloud server with substantial resources to train learning model. However, in this outsourcing paradigm it is of vital significance to address the privacy concern of client's data. Many researchers have been focusing on preserving the privacy of client's data in learning model. Recently, Wang et al. presented a privacy-preserving single-layer perceptron learning for e-healthcare scheme with using paillier cryptosystem and claimed that their scheme can protect the privacy of user's medical information. By analysing the process of iteration and the communication between the cloud and the user, we present that the honest-but-curious cloud can obtain the private medical information. Besides, the leakage of medical cases will lead to the exposure of the specific single-layer perceptron model of e-healthcare, which has gigantic commercial value.