Kernel interpolation-based technique for privacy protection of pluggable data in cloud computing

Abstract

Recent technological revolution has been reported as the revolution of the cloud computing technology. The explosive availability of the unstructured data in the cloud has gained the attention of the researchers and the users store their data in the cloud without any right over controlling the data, causing the privacy concerns. Therefore, there is a need for the effective privacy protection techniques that assure the privacy of the user data in the cloud. Accordingly, this paper proposes a kernel interpolation-based technique for preserving the privacy of the data in the cloud. Privacy and accuracy are the two factors assuring the privacy for the data, which are afforded using the proposed technique that uses the proposed rider-cat swarm optimisation (R-CSO) algorithm for computing the kernel interpolation coefficient, which is associated with affording the privacy in the cloud. The proposed rider-cat swarm optimisation (R-CSO) algorithm is the integration of the standard cat swarm optimisation (CSO) in the standard rider optimisation algorithm (ROA). The analysis of the methods using the dataset from UCI machine learning repository reveals that the proposed method acquired the maximal accuracy and minimal database difference ratio (DBDR) of 80.552% and 15.58%, respectively.