Ensure Data Privacy in Back Propagation Neural Network Learning over Encrypted Cloud Data | Abstract

ISSN ONLINE(2320-9801) PRINT (2320-9798)

Research Article Open Access

Ensure Data Privacy in Back Propagation Neural Network Learning over Encrypted Cloud Data

Abstract

Computational resources and storage resources are shared under the cloud environment through the Internet. In cloud environment users‟ data are usually processed remotely in unknown machines that users do not own or operate. Neural network techniques are used for the classification process. Collaborative Back-Propagation Neural Network (BPNN) learning is applied over arbitrarily partitioned data. The participating parties and the cloud servers are involved in the privacy preserved mining process. Each participant first encrypts their private data and then uploads the cipher texts to the cloud. Cloud servers execute most of the operations in the learning process over the cipher texts. Secure scalar product and addition operations are used in the encryption and decryption process. The collaborative learning process is handled without the Trusted Authority (TA). Key generation and issue operations are carried out in a distributed manner. Cloud server is enhanced to verify the user and data level details. Privacy preserved BPNN learning process is tuned with cloud resource allocation process.

M.T.Kiruthika, C.Selvi

To read the full article Download Full Article