A privacy-preserving high-order neuro-fuzzy c-means algorithm with cloud computing

Peng Li, Zhikui Chen, Laurence T. Yang, Liang Zhao, Qingchen Zhang

Abstract: Currently, massive heterogeneous data is generating from the Internet of Things (IoT). Heterogeneous data processing with the neuro-fuzzy technology has become a hot topic for IoT. In this work, we propose a privacy-preserving high-order neuro-fuzzy c-means algorithm for clustering heterogeneous data (PPHOFCM) on cloud computing. PPHOFCM clusters the heterogeneous data set by representing each heterogeneous data object as a tensor and uses the tensor distance to capture the correlations in the high-order tensor space. Furthermore, the cloud computing is employed to improve the clustering efficiency for massive heterogeneous data from IoT. The BGV encryption scheme is used to protect the private data when performing the high-order neuro-fuzzy c-means algorithm on cloud computing. Experiments are conducted on two real IoT datasets to verify the proposed algorithm.