Data Anonymization for Privacy Preservation in Big Data

Abstract

Cloud computing provides capable ascendable IT edifice to provision numerous processing of a various big data applications in sectors such as healthcare and business. Mainly electronic health records data sets and in such applications generally contain privacy-sensitive data. The most popular technique for data privacy preservation is anonymizing the data through generalization. Proposal is to examine the issue against proximity privacy breaches for big data anonymization and try to recognize a scalable solution to this issue. Scalable clustering approach with two phase consisting of clustering algorithm and K-Anonymity scheme with Generalisation and suppression is intended to work on this problem. Design of the algorithms is done with MapReduce to increase high scalability by carrying out dataparallel execution in cloud. Wide-ranging researches on actual data sets substantiate that the method deliberately advances the competence of defensive proximity privacy breaks, the scalability and the efficiency of anonymization over existing methods. Anonymizing data sets through generalization to gratify some of the privacy attributes like k- Anonymity is a popularly-used type of privacy preserving methods. Currently, the gauge of data in numerous cloud surges extremely in agreement with the Big Data, making it a dare for frequently used tools to actually get, manage, and process large-scale data for a particular accepted time scale. Hence, it is a trial for prevailing anonymization approaches to attain privacy conservation for big data private information due to scalabilty issues.