A new secure data retrieval system based on ECDH and hierarchical clustering with Pearson correlation

Abstract

Data retrieval is termed as the process of identification and extraction of appropriate data from a database management system as per the query provided by an application or user. In recent days, great concern is provided to data security that draws considerable research in the field of cloud computing. Particularly, cryptographic primitives for ensuring data security are adopted with prompt optimization of the existing algorithms. Prevailing cryptographic methods transfer huge data to be unintelligible leading to a major challenge in secured and efficient data retrieval. Therefore, there is a need for effective data retrieval system for transferring the data from owner to user. In that case, this paper aims to develop an efficient data retrieval system thereby accessing a reliable and secured data transfer system between the data owner and data user. For this purpose, this research utilized the advantages of the elliptic-curve Diffie–Hellman algorithm to generate encryption and decryption keys. Meanwhile, the proposed hierarchical clustering based on the Pearson correlation coefficient is employed for data segregation and extraction in accordance with the query. Here, the Pearson correlation coefficient is used for computing the distance matrix that allows the hierarchical clustering method to overcome the prevailing limitations such as computational and time complexity. Finally, the proposed data retrieval system outshines the state-of-the-art method in terms of range query processing time (17.4 ms) and key generation time (encryption time and decryption time as 17.4 ms and 13.2 ms).