PRIVACY PRESERVING IN DATA STREAM USING SLIDING WINDOW METHOD

Data mining gets valuable knowledge from huge amounts of data. In latest, data streams are new type of data, which are completely different from existing static data. The property of data streams are: Data has timing preference; data distribution changes constantly with time; the amount of data is large; here data flows in and out is very quickly; and on the spot reply is necessary. Existing algorithm is designed for the static database. If the data changes, it would be compulsory to rescan the whole dataset, which takes to more computation time and providing late respond to the user. Here we studied the problem of privacy-preserving data mining and many techniques have been find. However, existing techniques for privacy-preserving data mining are designed for static databases and are not suitable for dynamic data. When need to perform computation at that time to providing privacy. So the privacy preservation problem of data streams mining is very big issue. Here we generate data stream using MOA and then provide privacy to this stream. We proposed algorithms which take over the existing process of data streams classification to achieve more privacy preservation and accuracy.