A novel hybrid dwarf-based Archimedes optimization (HDAO) algorithm for preserving secure data in a cloud computing environment: A novel hybrid dwarf-based Archimedes optimization (HDAO) algorithm for

• Published in: Soft computing (Berlin, Germany) Germany), 2024, Vol.28 (23), p.13371-13387
• Main Author: Alandjani, Gasim
• Format: Article
• Language: English
• Subjects: Application of Soft Computing; Artificial Intelligence; Computational Intelligence; Control; Engineering; Mathematical Logic and Foundations; Mechatronics; Robotics
• ISSN: 1432-7643; 1433-7479
• DOI: 10.1007/s00500-024-10322-z

In general, cloud security is capable of providing various information, applications, services, etc. using extensive policies and progressive technologies. On the other hand, loss of data, confidentiality breaches, and loss of control are the major challenges in cloud security and they need to be solved. Therefore, this paper proposes a novel hybrid dwarf-based Archimedes optimization (HDAO) algorithm for optimal key generation and to solve both data privacy and integrity issues. The HDAO is the integration of both the dwarf optimization algorithm and Archimedes optimization algorithm. The main goal of the optimization algorithm is to solve the problem and find the optimal solutions by maximizing the objective function which makes them suitable for privacy-preserving tasks. One of the main weaknesses of the AOA algorithm is getting stuck in the local optima due to the usage of gradient-based methods. This issue is overcome using the dwarf optimization algorithm via randomization and local search strategy to complement the gradient-based approach. By employing the HDAO algorithm, the data sanitization and restoration process is effectively executed using the cloud data. In addition to this, the key extraction plays a vital role in the process of data sanitization and restoration process in which the optimization is carried out by the proposed HDAO algorithm. This paper employed five different types of datasets namely the genome, protein, general text, English text, and rand128 to evaluate the effectiveness of the proposed approach. Also, the comparative analysis is made for various parameters namely encryption time, decryption time, and avalanche effect, which is evaluated to determine the efficiency of the system.