Dyadic product and crow lion algorithm based coefficient generation for privacy protection on cloud

Cloud systems are powerful computing resources used inevitably for data subscription and publication. Even though the cloud platform can handle the huge volume of data, privacy becomes a critical issue during data publishing. Hence, an effective technique for the privacy preservation of the data is required in the cloud computing environment. Accordingly, this paper proposes a technique for privacy protection using the dyadic product and an optimization algorithm. The privacy of the original database is protected by the construction of privacy preserved database using a dyadic square matrix obtained taking the dyadic product of two vectors, namely sensitive-utility (SU) coefficient and cumulative data key product. The selection of SU coefficient vector is based on the proposed (Crow search based Lion) C-Lion algorithm, which is designed by combining crow search algorithm with lion algorithm. The fitness of the proposed C-Lion algorithm is designed based on privacy and utility for the feasible selection of SU coefficient vector. The performance of the proposed privacy protection technique based on the C-Lion algorithm is evaluated using two factors, privacy, and utility. The experimental analysis shows that the proposed technique could attain the maximum utility of 0.909 with privacy 0.864 for the breast cancer dataset.