A Survey on Privacy Preservation in Data Mining

Data mining is most usually used technique to extract the unfamiliar patterns from large data sets . Any transmission information to third parties i.e they should satisfy the condition to preserve the privacy of the data . Data Stream mining is the process of extracting knowledge structures from continuous data records[5]. The main problem in stream data mining is the evolving data . Privacy preserving data mining (PPDM) deals with the privacy of the individuals data and also without loosing the utility(accuracy) of the data[16] . On one hand data is the important asset for business decision making and analyzing on it . At the same time on the other hand the same data has many privacy concerns that might prevent the data owners to share those information for data analysis .This privacy & accuracy measure can be achieved by data mining task - Clustering & Classification . An efficient and effective approach has been proposed which aims at the privacy of the sensitive information and obtaining data with minimal information loss [14] . By using the Min-Max Normalization and by adding noise to the original data which is used as the composite method to preserve the privacy of the data[18].