An Enhanced Data Confidentiality in Online Social Networks Using Quantum Key Management with a Blockchain Approach
Sandip A. Kahate, Atul D. Raut · 2023
The main aim of this paper is to propose malicious detection and prevention by providing the data confidentiality framework using deep learning models of Long Short-Term Memory (LSTM) combined with a Convolutional Neural Network (CNN) for online social networks with a blockchain approach. Basically, this paper started with an introduction to deep learning and blockchain technology, followed by a literature survey on malicious attack detection, prevention, and data confidentiality in online social networks. Further, an analysis of the blockchain-based data confidentiality layers is presented, followed by the proposed malicious prevention framework using deep learning for decentralized online social networks (dOSN). Then, the comparative role of the LSTM-CNN deep learning prevention system with the existing system. The accuracy percentage for earlier sentiment analysis techniques utilizing the same data was just 64.47% compared to the more famous behavior analysis. This study increased accuracy to 96.63%, so the prospects are quite good, also compared and proved the Quantum Key Management (QKM) algorithm is more lightweight than classical cryptography algorithms in execution speed in the field of malicious prevention and protection systems is discussed, followed by the results and discussions, and lastly, the paper is concluded.