Main Article Content
Abstract
Globally, a huge amount of data is collected and published on daily basis which include a lot of sensitive information about the users. Thus, the privacy of data is a biggest concern and it is important to maintain the secrecy of such information. The data collected from centralized servers is processed under various data mining techniques for getting useful information, analysis and decision making. The trust of the users isentirely dependent on maintaining the privacy which may be violated while performing mining techniques on such a big data. Thus it is mandatory to develop an effective technique for data mining which preserve the data as well as information. It is further required that the data quality remains as per standard for further classification and analysis purposes. This paper is an attempt to study various privacy issues with a precise and systematic review of existing techniques of privacy preservation (Anonymization and Differential Privacy) and to put forward a concept which results into nominal information loss and better utility of data.