Volume 18, Special Issue on Current Trends in Management and Information Technology, 2021
A Hybrid Cryptographic Model Using AES and RSA for Sensitive Data Privacy Preserving
Satish B Basapur, B.S. Shylaja and Venkatesh
Abstract
In the present scenario, big data is facing many challenges regarding the data privacy and data security. Nowadays, new laws and regulations like GDPR is required for companies to define privacy policies complying with the preferences of their users. This type of regulations prevents the disclosure of sensitive data of users, even if occurs accidentally. In this research, a hybrid cryptographic model based on AES and RSA is proposed to identify and mask the sensitive data to identify many threats to information confidentiality. A hybrid cryptography algorithm in the context of masking is proposed to effectively transfer big data through the cloud. The hybrid algorithm is created by combining more than one algorithm. This algorithm enables the user to select the data to be masked and encrypted. For protecting data stored in the cloud, the proposed hybrid algorithm includes RSA and AES. Along with these algorithms, the multilayer perceptron neural network is used for key generation and key exchange. A credit card client's dataset from the UCI machine learning repository is used for the evaluation. From the dataset the sensitive attributes are selected using the depth first search (DFS) technique. The sender encrypts the data using the RSA algorithm and a key to create masked data using public key. The AES algorithm is used to encrypt the RSA key. The encrypted key and masked data are then sent to the receiver. To decrypt the RSA key, the receiver uses the AES decryption. Finally, the decrypted RSA key is used by the receiver to translate the masked data back to the original data with private key. The proposed model obtained the overall accuracy of 95.23% accuracy and an average computational time of 300 nano secs.
Pages: 129-148
DOI: 10.14704/WEB/V18SI05/WEB18219
Keywords: Privacy Preserving, Data Security, AES, RSA, DFS, Key Generation and Exchange, Neural Network, Information Confidentiality.