Privacy Preservation in Data Mining

Data mining aims to retrieve a useful or knowledgeable data from the large data repository. But privacy preservation means publishing the knowledgeable data along with hiding/protecting the sensitive data (personal data).Concerning illustration the aggregate utilization of information mining, vast information from claiming private majority of the data (data) would recurrently gathered furthermore broke down. Such information hold numerous shopping habits, restorative history, criminal records, kudos records and so forth throughout this way, observing and stock arrangement of all instrumentation may be etc.  And these data are the important asset to the governments and business organization for conclusion making process and to deliver social benefits such as health research, nationwide security etc. This is mostly concern through government agencies, insurance companies, hospitals and others that have information they would like to issue to researchers. Every record takes a number of attributes which can be separated into 3 categories: I. Explicit identifiers which can identify people. II. Quasi identifier features whose values when taken can simply recognize entity’s characteristics. III. Sensitive attributes which are considered as private and need not be revealed. Different anonymization methods have been explored to guard the sensitive data of the people. Re-identification of the data does not deliver anonymity, as the out data too contain quasi identifiers can be used for re-identifying the information of people, thus dripping that data which is not planned to be revealed data leakage occurs by coordination of quasi identifiers and external knowledge. Thus, the main aim is to bind the expose risk towards an acceptable level. So, it can be reached by anonymizing the data previously discharging it.

Professor,

Deeksha Gupta

Bhagwan Arihant College of Engineering & Technology, Surat.