Abstract

Cost Effective Privacy Preserving of Intermediate Data Set in Cloud Storage

Author(s): V. Sarala* and P. Shanmuga Priya

Data intensive applications store their valuable intermediate datasets in cloud in order to save the cost of re-computing. This poses a risk on data privacy protection because malicious parties may deduce the private information of the original datasets by analyzing multiple intermediate datasets. This system is implemented based on the least frequent pattern mining algorithm to identify the least frequent table and thereby encrypting it. From the least frequent table the reference attribute between the data tables are found out and a privacy leakage constraint is applied to the intermediate datasets by calculating the severity of the data to identify the sensitive information. As the result in the most frequent table only the privacy sensitive column alone is encrypted. In addition to this, an automatic scheduling algorithm is proposed to maintain a log based tracking for frequent and infrequent usage of data under the time criteria.