A STUDY OF PRIVACY PRESERVATION IN DATA MINING

Abstract

In order to extract the appealing unfamiliar patterns from large data sets, the most commonly used technique is Data Mining. The condition to preserve the privacy of data should be satisfied during the transmission of data to the third parties. From the continuous data records, extracting the knowledge structures is defined as Data Stream mining. Emerging data is an important problem in the Stream data mining. Privacy of the individual data, without losing the accuracy is dealt with Privacy preserving data mining abbreviated as PPDM. For the business decision making and evaluation, data is an important aid and on the other hand, lot of privacy concerns arise which prevents the owners of the data to share the particular information for the purpose of Data analysis. The Data Mining task called Clustering and Classification helps to achieve this certainty and Privacy measure. A competent and efficient approach data access with least information loss has been proposed. It considers the usage of Min-Max normalization and the addition of noise to the actual data which is a compound method in order to conserve the data privacy