A novel model for enhancing cloud security and data deduplication using fuzzy and refraction learning based chimp optimization

Recently, the digitalization process generates an enormous amount of multimedia data that turn out to be further difficult to manage. The current developments in big data technology and the cloud computing (CC) field produce massive growth in cloud data. The accessible memory space was used by the enormous replication data and generates the highest computation cost which is the most important problem in the constrained cloud storage space. Therefore, in this study, we introduced a novel secure cloud data deduplication (SCDD) approach to improve data security and data storage of the cloud environment by generating optimal key and deduplicating files respectively. The proposed approach mainly focuses on reducing computational cost and memory utilization of the cloud application. Here, the utilized data files are encrypted using the proxy re-encryption approach, and the refraction learning-based chimp optimization (RL-CO) algorithm is utilized for optimal key generation process as a result it guarantees better cloud security to the end users to store data on the cloud. Subsequently, the optimal verified fuzzy keyword search (OVFKS) approach is proposed to eliminate duplicate files or copies of actual data thereby enhancing the cloud storage space considerably. The proposed secure cloud data deduplication-based optimal verified fuzzy keyword search (SCDD-OVFKS)approach utilizes three different data files namely android application data, audio files, and mixed application data files, audio files, and other relevant files as the input. Furthermore, the proposed approach's performance is validated using different performance measures namely computational time, computational cost, search time cost, memory utilization, and deduplication rate by examining other state-of-art approaches. As a result, the proposed SCDD-OVFKS approach achieves a maximum deduplication rate of about 28.6% for 8 MB along with minimum computational cost and reduced memory utilization than other state-of-art approaches.